Syntheses of 2′-C-amidoalkyl and 2′-C-cyanoalkyl containing oligodeoxyribonucleotides and assessment of their hybridisation affinity for complementary DNA and RNA

Article abstract of DOI:10.1016/j.tet.2006.11.024

Oligodeoxynucleotides containing 2′-C-branched nucleosides with an amide or nitrile appended to either a one or two carbon alkyl chain have been synthesised. The phosphoramidites of the 2′-C-modified nucleosides were prepared and incorporated into the oli

Full text of DOI:10.1016/j.tet.2006.11.024

Tetrahedron 63 (2007) 577–585
Syntheses of 20-C-amidoalkyl and 20-C-cyanoalkyl containing
oligodeoxyribonucleotides and assessment of their hybridisation
affinity for complementary DNA and RNA
a
a
b
Lavinia Brennan,a, Richard Cosstick, Ian A. O’Neil and Arthur Van Aerschot
*
aRobert Robinson Laboratories, Department of Chemistry, University of Liverpool, PO Box 147, Liverpool L69 3BX, UK
bRega Institute for Medical Research, Laboratory for Medicinal Chemistry, 10 Minderbroedersstraat, Leuven 3000, Belgium
Received 7 June 2006; revised 25 October 2006; accepted 9 November 2006
Available online 29 November 2006
Abstract—Oligodeoxynucleotides containing 20-C-branched nucleosides with an amide or nitrile appended to either a one or two carbon
alkyl chain have been synthesised. The phosphoramidites of the 20-C-modified nucleosides were prepared and incorporated into the oligo-
nucleotides using automated DNA synthesis. The duplex stability with complementary RNA and DNA was measured by UV melting exper-
iments, in order to assess whether the amide/nitrile function could induce any duplex stability without the presence of the 20-oxygen. The
duplex stabilities of the oligonucleotides containing the 20-C-modifications were decreased in the absence of the 20-oxygen.
Ó 2006 Published by Elsevier Ltd.
1. Introduction
DNA-type conformation have generally been shown to cause
a decrease in duplex stability with complementary RNA.
Over the past decade a large number of nucleoside analogues
has been synthesised with the aim of enhancing the proper-
ties of antisense oligonucleotides.1–3 Ideally, a suitable
antisense candidate should have some resistance towards
nucleases, show increased duplex stability with complemen-
tary RNA4,5 and have the ability to invoke RNase H activity.6
Thus far, the most promising nucleic acid analogues for an-
tisense therapy include LNA7,8 and the 20-O-(methoxyethyl)
modification,9 both of which display some nuclease resis-
tance and an increased duplex stability with complementary
RNA. However, such analogues fail to activate RNase H.
The increased duplex stability displayed by these analogues
has been attributed to these modifications causing a shift in
the conformational equilibrium towards the C30-endo sugar
pucker. This shift is a result of the gauche effects between
the ring oxygen and the electronegative 20-O-substituent.
This gives rise to a structure, which is more RNA-like, and
thus forming stronger duplexes with complementary RNA.5
In order to activate RNase H, the antisense oligonucleotide is
required to be more DNA-like,10,11 having a predominantly
C20-endo sugar pucker. This conformation is favoured when
the 20-oxygen is replaced by a less electronegative atom such
as carbon. However, nucleoside analogues that adopt the
Previously we have investigated the synthesis of several
functionalised 20-C-branched nucleosides and encourag-
ingly, preliminary studies have shown that the incorporation
of these 20-C-modified residues into dinucleoside mono-
phosphates confers considerable resistance to nucleases.12
Related to this work, Sproat and co-workers have shown
that 20-O-(carbamoylmethyl) 1 containing oligonucleotides
show a higher affinity for RNA than those containing simple
20-O-alkyl or 20-O-allyl modifications.13 These results sug-
gest that in addition to the known beneficial effect of the
20-oxygen, the amide group itself has some stabilising ef-
fect on the duplex with complementary RNA. In order to
further investigate this, we report the synthesis of 20-C-
branched nucleosides containing an amide appended to
either a one 2 or two 3 carbon alkyl chain, followed by
incorporation into oligonucleotides and assessment of the
stability of duplexes formed with complementary RNA
and DNA. This study should give an indication whether
the effect of the amide alone is important for duplex stabil-
ity or if the increased affinity is reliant on the presence of
the 20-oxygen.
Another recent report in the literature investigated the effect
on binding affinity and nuclease resistance of 20-O-cyano-
ethyl modifications.14 Such modifications were found to be
stabilising and it was suggested that this was due to the
polarised nitrile bond inferring stability through H2O bridge
structures around the 20-position.
Keywords: 20-C-Modified nucleosides; Duplex stability; Antisense.
* Corresponding author at present address: Department of Chemistry,
University of Southampton, Southampton SO17 1BJ, UK. Tel.: +44 23
8059 6786; fax: +44 23 8059 2291; e-mail: lb15@soton.ac.uk
0040–4020/$ - see front matter Ó 2006 Published by Elsevier Ltd.
doi:10.1016/j.tet.2006.11.024

578
L. Brennan et al. / Tetrahedron 63 (2007) 577–585
5'
3'
5'
3'
5'
3'
B
B
B
O
O
O
O
O
O
O
O
O
2'
2'
2'
O
O
O
O
O
O
O
NH2
NH2
O P
O
O P
O
O P
O
NH2
B
B
B
5'
5'
O
O
5'
O
O
O
O
3
2
1
B = uracil-1-yl
In this paper we also report the syntheses and binding studies
of the 20-C-cyanoethyl and 20-C-cyanomethyl analogues and
comment on the comparison between these and the corre-
sponding 20-O-cyanoethyl analogue.
O
U
O
U
O
U
O
i
ii
O
O
TIPS
TIPS
TIPS
O
4
O
O
O
6
7
O
OMs
OH
iv
iii
2. Results and discussion
O
U
O
U
O
O
Previous attempts within this laboratory to prepare a 20-C-
amide-30-phosphoramidite building block 24 for incorpora-
tion into oligonucleotides had given low yields. It was also
observed that the amide group underwent dehydration in the
presence of the oxidant during oligonucleotide synthesis.12
TIPS
TIPS
O
CN
5
8
CN
Scheme 1. Preparation of 20-C-nitrile nucleosides. Reagents and conditions:
(i) NaBH4 in methanol, 99%; (ii) CH3SO2Cl in pyridine, 82%; (iii) NaCN in
DMSO, 71% and (iv) (a) hydroxylamine hydrochloride in pyridine/chloro-
form (b) selenium dioxide 66%.
It has been shown by Sproat and co-workers13 that oligo-
nucleotides containing a 20-O-nitrile can undergo aminolysis
of the 20-nitrile group by treatment with concentrated
aqueous ammonia. In addition, it is possible to carry out
this transformation under the standard conditions used for
removal of base and phosphate protecting groups and simul-
taneous cleavage of the oligonucleotide from the solid-phase
support. However, since completing this work, Sekine and
co-workers14 report that using a mixture of NH4OH and
NH4OAc for deprotection and cleavage leaves the nitrile
function intact.
The syntheses of the target 20-C-nitrile nucleosides 5 and 8
are illustrated in Scheme 1. Starting from the previously re-
ported 20-C-aldehyde12 4, 20-C-cyanomethyl nucleoside 5
was prepared using hydroxylamine hydrochloride followed
by treatment with selenium dioxide.15 Alcohol 6 was pre-
pared by simple sodium borohydride reduction in methanol
of the aldehyde 4. Initial attempts to access nitrile 8 via a di-
rect route using Mitsunobu conditions gave poor yields and
generated many side products making purification difficult.
It was found that accessing compound 8 via mesylate 7,
proved to be more facile and gave an overall higher yield.
It was envisaged that a 20-C-nitrile would be synthetically
more accessible and that this may avoid any potential prob-
lems previously associated with the 20-C-amide phosphor-
amidite. Post-synthetic treatment of the oligonucleotide
with aqueous ammonia was expected to result in aminolysis
of the 20-C-nitrile to give the 20-C-amide.
The 20-C-nitrile nucleosides 5 and 8 were prepared for
incorporation into oligonucleotides by conversion to the
corresponding 50-O-dimethoxytrityl-30-O-phosphoramidite
monomers. The synthesis is shown in Scheme 2.
DMTO
iii
U
DMTO
U
O
U
HO
U
O
O
O
O
ii
i
TIPS
R
O
P
OH
R
O
R
OH
R
N(iPr)2
CN(CH2)2O
5 R = CH2CN
8 R = (CH2)2CN
9 R = CH2CONH2
10 R = (CH2)2CONH2
11 R = CH2CO2H
12 R = CH2CN
18 R = CH2CN
22 R = CH2CN
13 R = (CH2)2CN
14 R = CH2CONH2
15 R = (CH2)2CONH2
16 R = CH2CO2H
19 R = (CH2)2CN
20 R = CH2CONH2
21 R = (CH2)2CONH2
23 R = (CH2)2CN
24 R = CH2CONH2
25 R = (CH2)2CONH2
Scheme 2. Preparation of 50-O-dimethoxytrityl-30-O-phosphoramidite nucleosides. Reagents and conditions: (i) NEt3$3HF in THF, 52–92%; (ii) dimethoxytrityl
chlorideinpyridine/DCM, 45–79%and (iii)2-cyanoethyl-bis-N,N,N,N-diisopropylaminophosphoramidite, diisopropylammoniumtetrazolide inCH3CN, 44–56%.

L. Brennan et al. / Tetrahedron 63 (2007) 577–585
579
The 20-C-nitrile-50-O-dimethoxytrityl-30-O-phosphorami-
dites 22 and 23 were incorporated into deoxyoligonucleo-
tides using standard automated solid-phase synthesis with
the phosphoramidite method.16–18 As previously mentioned,
it has been reported in the literature12 that by using the stan-
dard deprotection conditions (concentrated aqueous ammo-
nia at 55 ꢀC, 6 h is normal, although for convenience 16 h is
often used), a 20-O-nitrile undergoes aminolysis, to give the
corresponding 20-O-amide. Thus, we expected the same re-
action to happen with our 20-C-nitrile modification. Model
studies performed on the free 20-C-nitrile modified nucleo-
side had shown that when treated with aqueous ammonia
(16 h), the nitrile had been converted to a mixture of the
corresponding amide and carboxylic acid. Thus, to confirm
successful aminolysis of the nitrile groups, within the oligo-
nucleotides, the final deprotected oligonucleotides were
analysed by high-resolution mass spectrometry as well as
by enzyme digestion of the oligonucleotide to its constituent
nucleosides, followed by HPLC analysis. For the HPLC
analysis, standards of the 20-C-nitriles, 20-C-amide and 20-
C-carboxylic acids were required. The preparation of the
20-C-nitriles, 12 and 13 has already been discussed. The syn-
theses of the 20-C-carboxymethyluridine 16 and 20-C-amido-
methyluridine 14 are described in a previous publication.12
The syntheses of the standards 15 and 17 are shown below
in Scheme 3.
snake venom phosphodiesterase (SVPDE) and alkaline
phosphatase (AP). The resulting mixture of nucleosides was
analysed by reverse-phase HPLC, using conditions that had
been found to separate all the constituent nucleosides (see
Supplementary data). Previously within the group dinucleo-
tides containing either a 20-C-amide or 20-C-nitrile function
were found to have an increased resistance to hydrolysation
by SVPDE as compared to natural dinucleotides.12
The HPLC analysis of both oligonucleotides 27 and 28, con-
firmed that in both cases the nitrile functions had not under-
gone aminolysis. The transformation was also attempted
with methanolic ammonia, however, again no conversion
of the nitrile was evident. Initially it was hoped that employ-
ing the standard deprotection condition of 33% aqueous
ammonia at 55–60 ꢀC for 16 h would be sufficient enough
to transform the nitrile into an amide. However, this was
clearly not the case and the results were surprising, since pre-
liminary HPLC analysis of the monomer 20-C-nitrile had
shown evidence of partial aminolysis followed by complete
saponification upon treatment with 33% aqueous ammonia.
In addition, it had been previously reported that 20-O-nitrile,
which had been incorporated into an oligonucleotide and
treated in the same manner had in fact, undergone aminoly-
sis to give the 20-O-amide.
It is well documented that the presence of the carbon at the
20-position—as opposed to oxygen—gives the sugar pucker
of the modified nucleoside a predominantly C20-endo con-
formation due to loss of gauche interactions between O30
and O40. The difference in conformation is likely to change
the orientation of the nitrile group from that in the 20-O-
analogue. It may be noted that this change in orientation is
enough to prevent nucleophilic attack of ammonia, for ex-
ample, electrostatic repulsion may come from the anionic
oxygen of the phosphate linkage. Although problems were
envisaged with the incorporation of a 20-C-amide into an
oligonucleotide, it was decided to synthesise and directly in-
corporate the 20-C-amide-50-O-dimethoxytrityl-30-O-phos-
phoramidites. Thus the compounds 24 and 25 were
synthesised as shown in Scheme 2.
O
U
O
U
HO
U
i
ii
O
O
O
TIPS
TIPS
O
O
8
OH
10
13
CN
CN
H2N
O
iii
ii
HO
U
HO
U
O
O
HO
HO
17
15
HO
O
H2N
O
Scheme 3. Preparation of standards for HPLC. Reagents and conditions: (i)
MnO2/SiO2 in toluene, 40%; (ii)NEt3$3HFinTHF, 92%and(iii) 2 M NaOH.
Once the 20-C-phosphoramidites had been successfully
prepared, incorporation into the oligonucleotide of the
same sequence (30-CCGTUTAAATCC-50) as used for the
20-C-nitriles was attempted, using automated solid-phase
synthesis as before. The incorporation of the 20-C-amido-
ethyl 25 proved to be successful, although the trityl assays
showed that the coupling efficiency was quite low. How-
ever, coupling of the 20-C-methylamide proved to be
unsuccessful.
The purified unmodified oligonucleotide 26, together with
the oligonucleotides prepared with the phosphoramidites
containing the 20-C-nitrile modification 27 and 28 were char-
acterised by high-resolution mass spectrometry (Table 1). In
the case of oligonucleotides 27 and 28, the mass spectrum
data indicates that the nitrile function remains intact during
the extended deprotection and does not undergo aminolysis.
To confirm the presence of the 20-C-nitrile-containing nucleo-
sides, oligonucleotides 27 and 28 were then treated with
Previous attempts had been made to couple the 20-C-amido-
methyl-phosphoramidite 24 into a dinucleotide using solu-
tion-phase chemistry. This study had shown that during the
oxidation step, partial dehydration of the 20-C-amide had re-
sulted in the 20-C-nitrile.12 A mechanism for this reaction
had been proposed whereby a seven-membered ring inter-
mediate had been formed in the presence of an activator
such as tetrazole or DCI. Such a side reaction could clearly
interfere with the coupling of the 20-C-amidomethyl-30-O-
phosphoramidite 24, and could therefore account for the
low yield obtained.
Table 1. Characterisation of oligonucleotides by electrospray mass spec-
trometry
Oligo No. DNA sequence
Mass calcd Mass found
26
27
28
29
50-CCTAAAT-U-TGCC-30
3564.6
50-CCTAAAT-X1-TGCC-30 3603.6
50-CCTAAAT-X2-TGCC-30 3617.6
50-CCTAAAT-X3-TGCC-30 3635.7
3564.5
3603.6
3617.8
3635.8
26 U¼20-deoxyuridine, 27 X1¼20-CH2CN, 28 X2¼20-CH2CH2CN, 28
X3¼20-CH2CH2CONH2. Oligonucleotides 27, 28 and 29 result from the
incorporation of phosphoramidites 22, 23 and 25, respectively.

580
L. Brennan et al. / Tetrahedron 63 (2007) 577–585
A similar mechanism can be proposed for the 20-C-amido-
ethyl-30-phosphoramidite nucleoside, however, in this case
intramolecular cyclisation would lead to the less stable
eight-membered ring and would therefore be less favoured.
Nevertheless it cannot be ruled out since it would explain
why the coupling efficiency for the incorporation of the
20-C-amidoethyl nucleoside was lower than expected. The
fact that the 20-C-cyanoethyl coupled well, indicates that
factors other than steric hindrance are involved.
Some modifications such as the 20-O-methoxyethyl9 are
known to result in a significantly higher duplex stability,
thought to be due to an extended gauche effect between
O40–20-O and the oxygen of the methoxyethyl chain. This
20-substituent is also thought to aid hydration, adding further
stabilisation.24 Given that the 20-O-amide and 20-O-nitrile
have also resulted in a significant increase in duplex stability,
there could be a possibility that of such modifications to con-
tribute to such an extended gauche effect or similar stabilis-
ing effect due to an increase in hydration.
Duplex formation with the modified sequences was exam-
ined by mixing each oligomer with its complementary
DNA or RNA strand and determining the melting tempera-
tures (Tm)19 of the hybrids by temperature-dependent UV
spectrophotometry. Table 2 shows the Tm after incorporation
of the single modification X1, X2 or X3 into DNA oligomers.
3. Experimental
Most of the general analytical procedures have been de-
scribed in an earlier publication.12 NMR spectra were re-
corded at the field strength indicated, and peaks displaying
a multiplicity due to the presence of two diastereoisomers
are denoted with an asterisk. HPLC was conducted using
a Gilson HPLC, equipped with a Gilson 321 pump, Gilson
170 diode array detector, recording at 260 nm and a Gilson
234 auto injector. Operated using Unipoint version 3.0 soft-
ware. Unless otherwise stated, analyses were performed on
a Hichrom (250ꢁ4.6 mm) column packed with kromasil
100-C18. Eluent gradient employed A¼50 mM potassium
phosphate (pH 7), B¼60% A/40% CH3CN.
Table 2. Sequences (50-CCTAAATXTGCC-30, 30-GGATTTAAACGG-50)
and stability studies of the different duplexes carrying a single sugar modi-
fication
DNA complement
RNA complement
(0.1 M)
0.1 M
1.0 M
26
27
28
29
46.0
32.5
33.7
34.4
54.4
41.4
42.0
42.8
39.0
30.5
31.6
31.4
Snake venom phosphodiesterase (Crotalus adamanteus) was
purchased as lyophilised powder containing w35% Tris
buffer salts from Sigma (0.5 units) dissolved in 100 ml of
Tris buffer prior to use. Alkaline phosphatase (Bovine Intes-
tinal Mucosa) was purchased as a solution (1,000 DEA units
containing 50% glycerol, pH 7). Dissolved in 80 ml of 50:50
glycerol/water prior to use.
Tm values were determined at 260 nm in KH2PO4 buffer (20 mM, pH 7.5)
containing EDTA (0.1 mM) and NaCl (either 0.1 M or 1.0 M), with 4 mM
of each strand. Oligo 26, X¼U; Oligo 27, X¼20-CH2CN; 28, X¼20-
CH2CH2CN and 29 X¼20-CH2CH2CONH2.
The data show that, oligonucleotides 27–29 incorporating
the 20-C-modifications, result in a decrease in duplex stabil-
ity with complementary DNA and RNA as compared to the
unmodified sample 26. This is consistent with other studies,
which have shown that in general 20-C-modifications lead to
a lower Tm. This is known to be as a result of the absence of
the 20-O (or similar electronegative group), which is able to
adopt a favourable C30-endo conformation as a result of
gauche interaction with O40. In the DNA–DNA duplex at
low salt concentration the amidoethyl modification 29 gives
a slightly higher Tm than the corresponding nitrile 28. Inter-
estingly, comparing samples 27 and 28 the longer chain
length does not seem to have a destabilising effect than for
the same modification with one carbon less. This is not in
agreement with previous studies on 20-O-alkyl modifica-
tions, which have shown an inverse relationship between
chain length and binding affinity for complementary
RNA.20 The oligos 26–29 have higher duplex stabilities
with their DNA complement than with their RNA comple-
ment. Generally the stability of duplexes is in the order
RNA/RNA>DNA/RNA>DNA/DNA,21 however, this can
be sequence dependent. Often oligonucleotides containing
AT rich stretches—as in this case and others22,23—have
been found to show greater stability with their DNA comple-
ment than with their RNA complement.
3.1. Synthesis of monomers
3.1.1. 20-Deoxy-20-a-C-(2-oxoethyl)-30,50-O-(1,1,3,3,-
tetraisopropyldisiloxy)uridine (4). As described in previ-
ous publication.12 81%; 1H NMR d (200 MHz, CDCl3)
i
0.92–1.08 (28H, m, 4ꢁ Pr), 2.67 (1H, m, H60), 2.91–3.03
(2H, m, H20, H600), 3.90–3.98 (1H, m, H40), 4.03–4.97
(2H, m, H50, H500), 4.55 (1H, t, H30), 5.72–5.76 (2H, m,
H10, H5), 7.65 (1H, d, J6–5¼8.24 Hz), 9.55 (1H, br s, NH),
9.80 (1H, s, H700). 13C NMR d (75 MHz, CDCl3) 12.63–
13.31 (CH(CH3)2ꢁ4), 16.84–17.94 (CH3ꢁ8), 40.38 (C60),
43.48 (C20), 61.20 (C50), 69.51 (C30), 84.03 (C40), 88.49
(C10), 102.33 (C5), 139.14 (C6), 150.32 (C2), 163.08 (C4),
194.46 (C70). Found HRMS m/z (FAB+) [M+H]+
513.2451, C23H41N2O7Si2 requires 513.2452. Found C,
53.76; H, 7.93; N, 5.38. C23H40N2O7Si2 requires C, 53.88;
H, 7.87; N, 5.47.
3.1.2. 20-Deoxy-20-a-C-(cyanomethyl)-30,50-O-(1,1,3,3,-
tetraisopropyldisiloxy)uridine (5). Compound 4 (800 mg;
1.56 mmol) was dissolved in chloroform (w3 ml). To this
was added pyridine (1 equiv; 1.56 mmol; 0.13 ml) drop
wise with stirring. Hydroxylamine hydrochloride was added
(1.1 equiv; 1.72 mmol; 119 mg) and the mixture heated to
reflux. After 3 h TLC [hexane/EtOAc (50%)] showed no
starting material present. SeO2 (1 equiv; 1.56 mmol; 173 mg)
was added and the reaction was left to cool to room temper-
ature and then stirred overnight. After which time TLC
These results are in agreement with other studies, which
have shown that 20-O atom is essential for an increase in
duplex stability and highlight that either an amide group or
a nitrile group alone seem to have no significant effect.

L. Brennan et al. / Tetrahedron 63 (2007) 577–585
581
[hexane/EtOAc (50%)] showed the reaction to be complete.
Solvent was removed in vacuo and EtOAc (50 ml) was
added, this was washed with NaHCO3 (2ꢁ50 ml) and
NaCl (2ꢁ50 ml). All organic layers were dried over
MgSO4. Solvent was removed in vacuo to give crude product
as a yellow foam, which was then purified by column chro-
matography on silica gel, eluting with [hexane/EtOAc
(50%)] to give product 5 as a white foam (523 mg; 66%).
S–CH3), 3.85–3.93 (1H, m, H40), 4.00 (1H, d, H50), 4.18
(1H, dd, H500), 4.42–4.49 (3H, m, H30, H70, H700), 5.70
(1H, d, H5, J5–6¼8.24 Hz), 5.75 (1H, s, H10), 7.82 (1H, d,
H6, J6–5¼8.24 Hz), 9.37 (1H, br s, NH). 13C NMR
d (75 MHz, CDCl3) 12.55–13.37 (CH(CH3)2ꢁ4), 16.96
(CH3ꢁ8), 25.95 (C60), 37.35 (C20), 45.21 (CH3–S), 60.11
(C50), 67.95 (C30), 68.13 (C70), 83.13 (C40), 88.74 (C10),
101.94 (C5), 139.38 (C6), 150.37 (C2), 163.40 (C4). Found
HRMS m/z (FAB+) [M+H]+ 593.2375, C24H45N2O9SSi2
requires 593.2384.
i
1H NMR d (200 MHz, CDCl3) 1.00–1.08 (28H, m, 4ꢁ Pr),
2.72–2.75 (3H, m, H20, H60, H600), 4.05 (3H, m, H40, H50,
H500), 4.58 (1H, m, H30), 5.75 (1H, d, H5, J5–6¼7.7 Hz,
1H, d, H10), 7.67 (1H, d, H6, J6–5¼8.24 Hz). 13C NMR
d (75 MHz, CDCl3) 13.04–13.75 (CH(CH3)2ꢁ4), 15.75
(C60), 17.18–17.79 (CH3ꢁ8), 45.86 (C20), 61.06 (C50),
69.13 (C30), 83.87 (C40), 88.64 (C10), 102.66 (C5), 118.18
(CN), 139.21 (C6), 150.64 (C2), 163.62 (C4). Found
HRMS m/z [M+H+], 510.2447, C23H40N3O6Si2 requires
510.2455.
3.1.5. 20-Deoxy-20-a-C-(2-cyanoethyl)-30,50-O-(1,1,3,3,-
tetraisopropyldisiloxy)uridine (8). Compound 7 (315 mg;
0.53 mmol) was dissolved in the minimum amount of dry
DMSO (w3 ml). The solution was stirred under argon and
heated to 60 ꢀC then NaCN (2 equiv; 1.06 mmol; 52 mg)
was added and the mixture heated to 80 ꢀC. After 15 min
TLC [hexane/EtOAc (40%)] revealed the reaction to be com-
plete. The reaction was allowed to cool and EtOAc (30 ml)
was added, this was washed with NaHCO3 (2ꢁ20 ml)
and NaCl (2ꢁ20 ml). All organic layers were dried over
MgSO4. Solvent was removed in vacuo to give crude product
as a yellow oil, which was then purified by column chroma-
tography on silica gel, eluting with [hexane/EtOAc (0/
40%)] to give product 8 as a white foam (200 mg; 71%).
3.1.3. 20-Deoxy-20-a-C-(2-hydroxyethyl)-30,50-O-(1,1,3,3,-
tetraisopropyldisiloxy)uridine (6). Compound 4 (4.46 g;
8.71 mmol) was dissolved in MeOH (50 ml) and the solution
was cooled to 0 ꢀC. NaBH4 (3 equiv; 26.31 mmol; 0.99 g)
was added in small portions. The mixture was stirred for
25 min at 0 ꢀC, after this time TLC [DCM/MeOH (10%)]
showed the reaction to be complete. The reaction was
quenched by addition of citric acid crystals until neutral
pH was reached. Solvent was removed in vacuo to give
a white foam (4.44 g; 99%). 1H NMR revealed no further pu-
rification was required. 1H NMR d (200 MHz, CDCl3) 0.99–
i
1H NMR d (300 MHz, CDCl3) 0.79–0.88 (28H, m, 4ꢁ Pr),
1.56–1.62 (2H, m, H70, H700), 2.00–2.11 (2H, m, H60,
H600), 2.38–2.57 (1H, m, H20), 3.65 (1H, d, H40, J4 –5
¼
0
0
8.7 Hz), 3.77 (1H, dd, H50, J5 –5 ¼13.5 Hz, J5 -4 ¼8.7 Hz),
0
00
0
0
3.98 (1H, d, H500 J5 –5 ¼13.2 Hz), 4.27 (1H, t, H3 ), 5.46
(1H, s, H10), 5.59 (1H, d, H5, J5–6¼8.1 Hz), 7.60 (1H, d,
H6, J6–5¼8.1 Hz), 8.86 (1H, br s, NH). 13C NMR
d (75 MHz, CDCl3) 12.56–13.38 (CH(CH3)2ꢁ4), 15.46
(C70), 16.83–17.44 (CH3ꢁ8), 22.44 (C60), 47.89 (C20),
59.91 (C50), 67.90 (C30), 83.02 (C40), 88.48 (C10), 101.95
(C5), 119.60 (CN), 139.12 (C6), 150.16 (C2), 163.08 (C4).
Found HRMS m/z [M+H+], 524.2608, C24H42N3O6Si2
requires 524.2612. IR (CH2Cl2) 3398 cmꢂ1 (NH),
2872 cmꢂ1 (CH), 2250 cmꢂ1 (CN), 1695 cmꢂ1 (CO).
0
00
0
i
1.09 (28H, m, 4ꢁ Pr), 1.66 (1H, m, H60), 2.14 (1H, m, H600),
2.46 (1H, m, H20), 3.88–3.93 (3H, m, H40, H70, H700), 3.99
(1H, m, H50), 4.22 (1H, d, H500), 4.43 (1H, t, H50), 5.72
(1H, d, H5, J5–6¼8.2 Hz), 5.87 (1H, s, H10), 7.98 (1H, d,
H6, J6–5¼7.96 Hz), 9.25 (1H, br s, NH). 13C NMR
d (75 MHz, CDCl3) 12.87–13.77 (CH(CH3)2ꢁ4), 17.32–
17.92 (CH3ꢁ8), 28.18 (C60), 47.17 (C20), 60.10 (C70),
61.07 (C50), 67.57 (C30), 83.38 (C40), 89.39 (C10), 102.15
(C5), 140.04 (C6), 151.32 (C2), 163.91 (C4). Found
HRMS m/z (FAB+) [M+H]+ 515.2606, C23H43N2O7Si2
requires 515.2609. Found C, 53.64; H, 8.25; N, 5.38.
C23H42N2O7Si2 requires C, 53.67; H, 8.23; N, 5.45.
3.1.6. 20-Deoxy-20-a-C-(amidomethyl)-30,50-O-(1,1,3,3,-
tetraisopropyldisiloxy)uridine (9). As described in previ-
ous publication.12 1H NMR d (400 MHz, CDCl3) 1.01–
i
3.1.4. 20-Deoxy-20-a-C-(2-methylsulfonyloxyethyl)-30,50-
O-(1,1,3,3,-tetraisopropyldisiloxy)uridine (7). Compound
6 (4.44 g; 8.63 mmol) was co-evaporated with dry pyridine
(2ꢁ10 ml) and then dissolved in dry pyridine (50 ml) and
cooled to 0 ꢀC under argon. Mesyl chloride (2 equiv;
12.27 mmol; 1.34 ml) was added drop wise and the solution
was stirred at room temperature for 1 h (by which time the
solution had become dark brown in colour). After this time
TLC [DCM/MeOH (10%)] showed the reaction to be com-
plete. A few drops of H2O were added to quench the reaction
and then the pyridine was removed in vacuo by co-evapora-
tion with toluene. The resulting residue was taken up in
DCM (50 ml) and washed with NaHCO3 (2ꢁ50 ml) and
NaCl (2ꢁ50 ml). All organic layers were dried over
MgSO4. Solvent was removed in vacuo to give crude product
as an orange oil, which was then purified by column chroma-
tography on silica gel, eluting with [DCM/MeOH (0/3%)]
1.10 (28H, m, 4ꢁ Pr), 2.47 (1H, dd, H60, J6 –2 ¼10.34 Hz,
0
0
00 0
0
00
J6 –6 ¼17.4 Hz), 2.63–2.73 (2H, m, H2 , H6 ), 3.90 (1H,
dd, H50, J5 –4 ¼7.3 Hz, J5 –5 ¼11.6 Hz), 3.97 (1H, m, H4 ),
0
0
0
0
00
4.08 (1H, dd, H500, J5 –4 ¼3.4 Hz, J5 –5 ¼11.5 Hz), 4.53
00
0
00
0
(1H, dd, H30, J3 –4 ¼5.0 Hz, J3 –2 ¼7.1 Hz), 5.75 (1H, d,
00
0
0
0
H5, J5–6¼8.3 Hz), 5.95 (1H, d, H10, J1 –2 ¼6.1 Hz), 6.00
(1H, br s, NH-amide), 6.70 (1H, br s, NH-amide), 7.47
(1H, d, H6, J6–5¼8.1 Hz), 9.89 (1H, br s, NH-uracil). 13C
NMR d (75 MHz, CDCl3) 12.61–13.53 (CH(CH3)2ꢁ4),
16.92–17.53 (CH3ꢁ8), 31.74 (C60), 44.93 (C20), 62.84
(C50), 72.10 (C30), 85.30 (C40), 88.14 (C10), 103.15 (C5),
139.18 (C6), 151.40 (C2), 163.19 (C4), 173.88 (C70). Found
HRMS m/z [M+H+], 528.2557, C23H41N2O8Si2 requires
528.2561.
0
0
3.1.7. 20-Deoxy-20-a-C-(2-amidoethyl)-30,50-O-(1,1,3,3,-
tetraisopropyldisiloxy)uridine (10). Compound
8
1
to give product 7 as a white foam (4.2 g; 82%). H NMR
d (200 MHz, CDCl3) 1.00–1.08 (28H, m, 4ꢁ Pr), 1.88
(200 mg; 0.38 mmol) was dissolved in toluene (w5 ml) and
a freshly prepared sample of MnO2/SiO2$H2O* (9.2 equiv;
3.48 mmol; 4.2 g) was added. The mixture was heated to
i
(1H, m, H20), 2.37 (2H, m, H60, H600), 3.05 (3H, s,

582
L. Brennan et al. / Tetrahedron 63 (2007) 577–585
reflux and stirred under argon for 24 h. After this time, TLC
[DCM/MeOH (10%)] showed reaction to be complete. Sol-
vent was evaporated in vacuo and the residue was dissolved
in MeOH and filtered under suction and then re-filtered
under gravity to remove as much of the silica as possible.
Solvent was removed in vacuo to give crude product as
a white foam, which was then purified by column chromato-
graphy on silica gel, eluting with [DCM/MeOH (0/4%)] to
give product 10 as a white foam (81 mg; 40%). According to
Lui et al.25 each gram of supported reagent is found to con-
tain 0.83 mmol of MnO2 (as determined by titration against
sodium oxalate and potassium permanganate) and 3.1 g of
free H2O (as determined by Karl–Fischer method). 1H
142.49 (C6), 153.06 (C2), 166.41 (C4). Found HRMS m/z
[M+H+], 268.0933, C11H14N3O5 requires 268.0931.
3.2.2. 2-Deoxy-20-a-C-(2-cyanoethyl)uridine (13). 83%;
1H NMR d (400 MHz, MeOD) 1.73–1.77 (1H, m, H70),
2.05–2.10 (1H, m, H700), 2.36–2.44 (1H, m, H60), 2.54–
2.56 (2H, m, H20, H600), 3.37 (2H, m, H50, H500), 4.01 (1H,
m, H40), 4.34 (1H, d, H30, J3 –4 ¼4.92 Hz), 4.52 (1H, br s,
0
0
OH), 5.76 (1H, d, H5, J5–6¼8.08 Hz), 6.05 (1H, d, H10,
J1 –2 ¼8.56 Hz), 8.03 (1H, d, H6, J6–5¼7.96 Hz). 13C
NMR d (75 MHz, MeOD) 16.03 (C60), 22.15 (C70), 48.78
(C20), 63.66 (C50), 73.33 (C30), 89.33 (C40), 89.54 (C10),
103.79 (C5), 121.19 (CN), 142.73 (C6), 153.05 (C2),
166.38 (C4). Found HRMS m/z [M+H+], 282.1091,
C12H16N3O5 requires 282.1090.
0
0
i
NMR d (400 MHz, CDCl3) 1.00–1.09 (28H, m, 4ꢁ Pr),
1.71 (2H, m, H70, H700), 2.17 (1H, m, H60), 2.25 (1H, m,
H600), 2.56 (1H, m, H20), 2.70 (1H, 0m0 , H40), 3.94 (1H, d,
H50, J5 –5 ¼16 Hz), 4.21 (1H, d, H5 , J5 –5 ¼16 Hz), 4.42
3.2.3. 2-Deoxy-20-a-C-(amidomethyl)uridine12 (14). 92%;
1H NMR d (200 MHz, MeOD) 2.43 (1H, m, H60), 2.68 (2H,
0
00
00
0
(1H, t, H30), 5.57 (1H, br s, NH), 5.59 (1H, d, H5, J5–6
¼
8 Hz), 5.74 (1H, s, H10), 6.68 (1H, br s, NH), 7.96 (1H,
d, H6, J6–5¼8 Hz), 10.68 (1H, br s, NH). 13C NMR
d (75 MHz, CDCl3) 12.05–13.38 (CH(CH3)2ꢁ4), 16.44–
17.08 (CH3ꢁ8), 19.72 (C60), 32.10 (C70), 48.37 (C20),
59.44 (C50), 66.99 (C30), 82.52 (C40), 8.60 (C10), 101.75
(C5), 139.47 (C6), 151.37 (C2), 163.61 (C4), 175.62 (C80).
Found HRMS m/z [M+Na+], 564.25542, C24H43N3O7Si2Na
requires 564.25373.
m, H20, H600), 3.75 (2H, d, H50, H500, J5 –4 ¼3.3 Hz), 3.99
0
0
(1H, m, H40), 4.34 (1H, m, H30), 5.72 (1H, d, H5,
J5–6¼8.24 Hz), 6.04 (1H, d, H10, J1 –2 ¼8.26 Hz), 7.98
0
0
(1H, d, H6 J6–5¼7.98 Hz).
3.2.4. 2-Deoxy-20-a-C-(2-amidoethyl)uridine (15). 83%;
1H NMR d (400 MHz, MeOD) 1.73–1.77 (1H, m, H70),
2.05–2.10 (1H, m, H700), 2.36–2.44 (1H, m, H60), 2.54–
2.56 (2H, m, H20, H600), 3.37 (2H, m, H50, H500), 4.01 (1H,
m, H40), 4.34 (1H, d, H30, J3 –4 ¼4.92 Hz), 4.52 (1H, br s,
3.1.8. 20-Deoxy-20-a-C-(carboxymethyl)-30,50-O-(1,1,3,3,-
tetraisopropyldisiloxy)uridine (11). As described in
0
0
OH), 5.76 (1H, d, H5, J5–6¼8.08 Hz), 6.05 (1H, d, H10,
J1 –2 ¼8.56 Hz), 8.03 (1H, d, H6, J6–5¼7.96 Hz). 13C
NMR d (75 MHz, MeOD) 16.03 (C60), 22.15 (C70), 48.78
(C20), 63.66 (C50), 73.33 (C30), 89.33 (C40), 89.54 (C10),
103.79 (C5), 121.19 (CN), 142.73 (C6), 153.05 (C2),
166.38 (C4). Found HRMS m/z [M+H+], 282.1091,
C12H16N3O5 requires 282.1090.
previous publication.12 1H NMR d (400 MHz, CDCl3)
0
0
i
1.02–1.10 (28H, m, 4ꢁ Pr), 2.46 (1H, dd, H60, J6 –2
¼
0
0
00
0
0
00
10.8 Hz, J6 –6 ¼14.9 Hz), 2.74–2.84 (2H, m, H2 , H6 ),
3.87 (1H, m, H40), 4.00–4.10 (2H, m, H50, H500), 4.50 (1H,
t, H30, J3 –2 ¼7.7 Hz, J3 –4 ¼3.9 Hz), 5.76 (1H, d, H5, J5–6
¼
0
0
0
0
8.1 Hz), 5.93 (1H, d, H10, J1 –2 ¼3.2 Hz), 7.65 (1H, d, H6,
J6–5¼8.1 Hz), 10.60 (1H, br s, NH). 13C NMR d (75 MHz,
CDCl3) 12.99–13.79 (CH(CH3)2ꢁ4), 17.24–17.81 (CH3ꢁ8),
31.56 (C60), 46.34 (C20), 61.32 (C50), 69.47 (C30), 84.04
(C40), 88.95 (C10), 102.96 (C5), 139.66 (C6), 151.51 (C2),
164.34 (C4). Found HRMS m/z [M+H+], 529.2396,
C23H42N2O8Si2 requires 529.2401.
0
0
3.2.5. 2-Deoxy-20-a-C-(carboxymethyl)uridine12 (16).
67%; 1H NMR d (400 MHz, MeOD) 2.37 (1H, dd,
00
H60, J6 –2 ¼6.52 Hz, J6 –6 ¼16 Hz), 2.65 (1H, dd, H6 ,
0
0
0
00
0
00
0
00
0
J6 –2 ¼7.80 Hz, J6 –6 ¼16 Hz), 2.72–2.76 (1H, m, H2 ),
3.76 (2H, d, H50, H500, J5 –4 ¼3.68 Hz), 4.00–4.02 (1H, m,
H40), 4.33 (1H, dd, H30, J¼1.28 Hz, J¼2 Hz), 4.52 (1H,
br s, OH), 5.73 (1H, d, H5, J5–6¼8.12 Hz), 6.07 (1H, d,
0
0
3.2. General procedure for the removal of the
1,1,3,3-tetraisopropyldisiloxy protecting group
H10, J1 –2 ¼8.5 Hz), 7.95 (1H, d, H6, J6–5¼8.08 Hz). 13C
NMR d (75 MHz, MeOD) 31.82 (C60), 47.06 (C20), 63.95
(C50), 74.69 (C30), 89.43 (C40), 89.88 (C10), 103.65 (C5),
143.05 (C6), 152.65 (C2), 166.50 (C4), 176.97 (C70).
0
0
The TIPS-protected nucleoside (0.45 mmol) was dissolved
in dry THF (30 ml), and to this was added NEt3$3HF
(0.08 mL, 5 mmol). The solution was left for 8–16 h and
monitored by TLC. The solution was concentrated in vacuo,
diluted with methanol (20 ml) and evaporated onto silica
gel; the product was subsequently isolated by column chro-
matography (DCM containing an increasing gradient of
MeOH from 0–10%) as a colourless oil or white foam.
3.2.6. 2-Deoxy-20-a-C-(2-carboxyethyl)uridine (17).
Compound 13 (10 mg) was dissolved in 2 M aqueous so-
dium hydroxide solution (1 ml) and stirred at room temper-
ature for 7 days. The reaction progress was followed by
reverse-phase HPLC. After one week no starting material re-
mained and HPLC revealed a single peak with a retention
time of 8.74 min. The product 17 was collected (w10 mg).
3.2.1. 20-Deoxy-20-a-C-(cyanomethyl)uridine (12). 52%;
1H NMR d (200 MHz, MeOD) 2.59–2.79 (3H, m, H20,
y
1H NMR d (400 MHz, MeOD) 1.08–1.14 (1H, m, H70),
1.30–1.38 (3ꢁ[HNCH2CH3]+), 1.63 (1H, br s, H700), 1.98–
H60, H600), 3.75 (2H, d, H50, H500, J5 –4 ¼3.28 Hz), 4.03
2.27 (3H, m, H60, H600, H20), 3.17–3.21 (3ꢁ[HNCH2CH3]+),
0
0
(1H, m, H40), 4.32 (1H, m, H30), 5.74 (1H, d, H5,
J5–6¼8.24 Hz), 6.07 (1H, d, H10, J1 –2 ¼8.24 Hz), 7.97
(1H, d, H6, J6–5¼8.24 Hz). 13C NMR d (100 MHz,
MeOD) 14.11 (C60), 47.13 (C20), 63.55 (C50), 73.93 (C30),
89.27 (C40), 89.47 (C10), 103.85 (C5), 119.96 (CN),
0
0
Product was characterised by accurate mass spectrometry and 1H NMR.
13C NMR data were also collected, however due to the weak sample con-
centration quaternary carbons were not seen on the spectra and therefore
only selected data are reported.
y

L. Brennan et al. / Tetrahedron 63 (2007) 577–585
583
0
3.73 (2H, d, H50, H500, J5 –4 ¼2.16 Hz), 4.00 (1H, s, H4 ),
4.30 (1H, s, H30), 5.74 (1H, d, H5, J5–6¼6.42 Hz), 6.04
(1H, s, H10), 7.97 (1H, d, H6, J6–5¼6.70 Hz). 13C NMR
yd (75 MHz, MeOD) 31.11 (C60), 48.27 (C20), 50.80 (C70),
64.10 (C50), 73.95 (C30), 89.14 (C40), 90.00 (C10), 103.64
(C5), 143.07 (C6). Found HRMS m/z (FABꢂ) [MꢂH]+
299.08876, C12H14N2O7 requires 299.08793.
30-OH), 5.41 (1H, d, H5, J5–6¼8.08 Hz), 6.06 (1H, d, H10,
0
0
0
0
J1 –2 ¼7.44 Hz), 6.44 (1H, br s, NH-amide), 6.72 (1H, br s,
NH-amide), 6.81 (4H, d, ArH o-OCH3, J¼8.92 Hz), 7.16–
7.39 (9H, m, ArH), 7.64 (1H, d, H6, J6–5¼8.12 Hz), 10.37
(1H, br s, NH-uracil). 13C NMR d (100 MHz, CDCl3)
31.26 (C60), 46.85 (C20), 55.25 (OCH3), 63.80 (C50), 73.45
(C30), 85.96 (C40), 86.91 (OCAr3), 88.24 (C10), 102.97
(C5), 113.34 (ArC), 127.09 (ArC), 128.01 (ArC), 130.13
(ArC), 135.36 (ArC), 135.51 (ArC), 140.30 (C6), 144.36
(ArC), 151.74 (C2), 158.69 (ArC), 163.67 (C4), 175.02
(CONH2). Found HRMS m/z [M+Na+], 610.2137,
C32H33N3O78Na requires 610.2165.
3.3. General procedure for the preparation
of 50-O-dimethoxytrityl ethers
The nucleoside (150 mg; 0.61 mmol) was dissolved in dry
pyridine (1.5 ml). To this, a solution of dimethoxytrityl chlo-
ride (1.7 equiv; 1.03 mmol; 349 mg) in dry pyridine and dry
DCM (1:3.5) (1 ml) was added drop wise over a period of
20 min. The solution was stirred under argon for 4 h, after
this time, TLC [DCM/MeOH (10%)] showed the reaction
to be complete. The reaction was quenched by addition of
methanol (w0.5 ml), followed by addition of NaHCO3
(0.5 ml). Solvent was removed in vacuo and residue was
taken up in EtOAc (w5 ml). The mixture was then washed
with NaHCO3 (2ꢁ5 ml) and NaCl (2ꢁ5 ml). All organic
layers were dried over MgSO4. Solvent was removed
invacuo to give crude product, which was purified by column
chromatography on silica gel, eluting with [DCM/MeOH
(0/3%)] to give the product as a white foam.
3.3.4. 20-Deoxy-20-a-C-(2-amidoethyl)-50-O-dimethoxy-
trityluridine (21). 70%; H NMR d (400 MHz, CDCl3)
1
1.62–1.66 (1H, m, H60), 2.03 (1H, br s, H600), 2.21–2.25
(1H, m, H20), 2.35–2.46 (H, m, H70, H700), 3.40 (2H, s,
2ꢁH50), 3.74 (6H, s, OCH3), 4.10 (1H, d, H40, J4 –5
¼
0
0
2.72 Hz), 4.38 (1H, br s, H30), 4.58 (1H, br s, OH), 5.34
(1H, d, H5, J5–6¼8.12 Hz), 5.99 (1H, d, H10, J1 –2
¼
0
0
7.28 Hz), 6.19 (1H, br s, NH-amide), 6.44 (1H, br s, NH-
amide), 6.81 (4H, d, ArH o-OCH3, J¼8.88 Hz), 7.16–7.36
(9H, m, ArH), 7.75 (1H, d, H6, J6–5¼8.08 Hz),w10 (1H, br
13
s, NH-uracil). C NMR d (100 MHz, CDCl3) 31.25 (C60),
33.57 (C20), 50.38 (C70), 55.62 (OCH3), 64.00 (C50), 72.23
(C30), 85.64 (C40), 87.40 (OCAr3), 88.66 (C10), 103.12 (C5),
113.76 (ArC), 127.48 (ArC), 128.36 (ArC), 130.50 (ArC),
135.65 (ArC), 135.84 (ArC), 140.84 (C6), 144.75 (ArC),
151.62 (C2), 159.06 (ArC), 164.02 (C4), 176.78 (CONH2).
3.3.1. 20-Deoxy-20-a-C-(cyanomethyl)-50-O-dimethoxy-
trityluridine (18). 58%; H NMR d (400 MHz, MeOD)
1
2.64 (2H, m, H20, H60), 2.82 (1H, m, H600), 3.47 (3H, m,
H40, 2ꢁH50), 3.71 (1H, m, H30), 3.79 (6H, s, OCH3),
3.4. General procedure for the preparation of nucleo-
side-30-[(2-cyanoethyl)-N,N-diisopropyl]phosphor-
amidites
5.48 (1H, d, H5, J5–6¼8.1 Hz), 6.07 (1H, d, H10, J1 –2
¼
0
0
7.6 Hz), 6.85 (4H, d, ArH o-OCH3, J¼7.1 Hz), 7.22–
7.40 (9H, m, ArH), 7.65 (1H, d, H6, J6–5¼8.3 Hz), 9.20
(1H, br s, NH). 13C NMR d (100 MHz, MeOD) 13.79
(C60), 46.80 (C20), 55.68 (OCH3), 63.66 (C50), 72.84 (C30),
86.53 (C40), 87.73 (OCAr3), 87.80 (C10), 103.55 (C5),
113.69 (ArC), 113.79 (ArC), 118.48 (CN), 127.68 (ArC),
128.43 (ArC), 128.52 (ArC), 130.43 (ArC), 130.46 (ArC),
135.36 (ArC), 139.84 (C6), 144.43 (ArC), 151.10 (C2),
159.16 (ArC), 163.25 (C4). Found HRMS m/z [M+H+],
570.2235, C33H33N3O7 requires 570.2240.
The 50-O-tritylated nucleoside (235 mg; 0.43 mmol) and
diisopropylammonium tetrazolide (0.5 equiv; 0.21 mmol;
36.39 mg) were dissolved in dry CH3CN (4 ml). To
this, 2-cyanoethyltetraisopropylphosphoramidite (1.1 equiv;
0.47 mmol; 0.15 ml) was added drop wise. The solution
was stirred under argon for 16 h, after this time, TLC
[DCM/MeOH (10%)] showed the reaction to be complete.
The reaction mixture was diluted with EtOAc (w20 ml)
and then washed with NaHCO3 (2ꢁ10 ml) and NaCl
(2ꢁ10 ml). All organic layers were dried over MgSO4. Sol-
vent was removed in vacuo to give crude product, which was
purified by column chromatography on silica gel, eluting
with [hexane/EtOAc/NEt3 (50:49:1%)] to give product as
a white oil. This was then precipitated from hexane to give
a white foam.
3.3.2. 20-Deoxy-20-a-C-(2-cyanoethyl)-50-O-dimethoxy-
trityluridine (19). 79%; H NMR d (400 MHz, MeOD)
1
1.77 (1H, m, H60), 2.15 (1H, m, H600), 2.53 (3H, m, H20,
H70, H700), 3.47 (2H, 2ꢁH50, J5 –4 ¼2.28 Hz), 3.79 (6H, s,
0
0
OCH3), 4.09 (1H, m, H40), 4.51 (1H, m, H30), 5.38 (1H, d,
H5, J5–6¼8.4 Hz), 6.11 (1H, d, H10, J1 –2 ¼7.84 Hz), 6.85
(4H, d, ArH o-OCH3, J¼8.8 Hz), 7.31 (9H, m, ArH), 7.78
(1H, d, H6, J6–5¼8.1 Hz), 9.18 (1H, br s, NH). 13C NMR
d (100 MHz, MeOD) 14.59 (C60), 15.87 (C70), 48.53 (C20),
55.67 (OCH3), 63.78 (C50), 72.67 (C30), 86.67 (C40), 87.72
(OCAr3), 88.23 (C10), 103.42 (C5), 113.77 (ArC), 119.73
(CN), 127.65 (ArC), 128.52 (ArC), 130.51 (ArC), 135.36
(ArC), 135.54 (ArC), 140.48 (C6), 144.44 (ArC), 151.19
(C2), 159.14 (ArC), 163.00 (C4). Found HRMS m/z
[M+NH+4], 601.2663, C33H41N5O7 requires 601.2662.
0
0
3.4.1. 20-Deoxy-20-a-C-20-(cyanomethyl)-50-O-dimethoxy-
trityluridine-[(2-cyanoethyl)-N,N-diisopropyl]phosphor-
amidite (22). 56%; 1H NMR d (400 MHz, MeOD)
1.11–1.31 (12H, m, 4ꢁCH3), 2.48–2.72 (5H, m, CH2CN,
H20, H60, H600), 3.42–3.71 (6H, m, 2ꢁMe2CHN,
NCCH2CH2O, 2ꢁH50), 3.81 (6H, s, OCH3), 4.29y, 4.39y
(1H, br s, H40), 4.51y, 4.68y (1H, m, H30) 5.44y, 5.50y
(1H, d, H5, J5–6¼8.2 Hz, J5–6¼8.1 Hz), 6.04y, 6.05y (1H,
3.3.3. 20-Deoxy-20-a-C-(amidomethyl)-50-O-dimethoxy-
trityluridine (20). 45%; H NMR d (400 MHz, CDCl3)
d, H10, J1 –2 ¼4.1 Hz, J1 –2 ¼3.7 Hz), 6.85–6.89 (4H, m,
ArH o-OCH3), 7.26–7.39 (9H, m, ArH), 7.72y, 7.69y (1H,
d, H6, J6–5¼8.1 Hz, J6–5¼8.1 Hz). 31P NMR d (161 MHz,
MeOD) 152.25, 153.47. Found HRMS m/z [M+H+],
770.3324, C41H50N5O8P requires 770.3319.
0
0
0
0
1
2.57–2.70 (3H, m, H20, H60, H600), 3.38 (2H, d, 2ꢁH50,
0
0
0
J5 –4 ¼2.68 Hz), 3.75 (6H, s, OCH3), 4.14 (1H, d, H4 ,
0
0
0
J4 –5 ¼2.36 Hz), 4.33 (1H, br s, H3 ), 4.45 (1H, br s,

584
L. Brennan et al. / Tetrahedron 63 (2007) 577–585
3.4.2. 20-Deoxy-20-a-C-20-(2-cyanoethyl)-50-O-dimethoxy-
trityluridine-[(2-cyanoethyl)-N,N-diisopropyl]phosphor-
amidite (23). 45%; 1H NMR d (400 MHz, MeOD)
1.11–1.33 (12H, m, 4ꢁCH3), 1.80 (1H, m, H60), 2.34 (1H,
m, H600), 2.48–2.76 (5H, m, CH2CN, H20, H70, H700), 3.40–
3.73 (6H, m, 2ꢁMe2CHN, NCCH2CH2O, 2ꢁH50), 3.79y,
3.80y (6H, s, OCH3), 4.19y, 4.30y (1H, br s, H40), 4.53y,
4.65y (1H, m, H30), 5.33y, 5.40y (1H, d, H5, J5–6¼8.2 Hz,
(A¼10 mM NaOH, pH 12.0, 0.1 M NaCl; B¼10 mM
NaOH, pH 12.0, 0.9 M NaCl). The low-pressure liquid chro-
matography system consisted of a Merck-Hitachi L 6200 A
intelligent pump, a Mono QÒ-HR 10/10 column (Pharma-
cia), a Uvicord SII 2138 UV detector (Pharmacia-LKB)
and a recorder. The product-containing fraction was desalted
on a NAP-25Ò column and lyophilised.
J5–6¼8.1 Hz), 6.04y, 6.06y (1H, d, H10, J1 –2 ¼6.6 Hz,
Oligonucleotides were characterised and their purity was
checked by HPLC–MS on a capillary chromatograph
(CapLC, Waters, Milford, MA). Columns of 150 mmꢁ
0.3 mm length (LCPackings, San Francisco, CA) were
used. Oligonucleotides were eluted with a triethylammo-
nium/1,1,1,3,3,3-hexafluoro-2-propanol/acetonitrile solvent
system. Flow rate was 5 ml/min. Electrospray spectra were
acquired on an orthogonal acceleration/time-of-flight mass
spectrometer (Q-Tof-2, Micromass, Manchester, UK) in neg-
ative ion mode. Scan time used was 2 s. The combined spec-
tra from a chromatographic peak were deconvoluted using
the MaxEnt algorithm of the software (Masslynx 3.4, Micro-
mass, Manchester, UK). Theoretical oligonucleotide masses
were calculated using the monoisotopic element masses.
0
0
0
0
J1 –2 ¼8.1 Hz), 6.84–6.89 (4H, m, ArH o-OCH3), 7.25–
7.39 (9H, m, ArH), 7.80y, 7.83y (1H, d, H6, J6–5¼8.3 Hz,
J6–5¼8.1 Hz), 9.30 (1H, br s, NH). 31P NMR d (161 MHz,
MeOD) 150.67, 152.73. Found HRMS m/z [M+H+],
784.3477, C42H52N5O8P requires 784.3475.
3.4.3. 20-Deoxy-20-a-C-20-(amidomethyl)-50-O-dimethoxy-
trityluridine-[(2-cyanoethyl)-N,N-diisopropyl]phosphor-
amidite (24). 52%; 1H NMR d (400 MHz, CDCl3) 1.12–1.29
(12H, m, 4ꢁCH3), 2.47–2.80 (4H, m, CH2CN, H20, H60),
2.96 (1H, m, H600), 3.37–3.88 (6H, m, 2ꢁMe2CHN,
NCCH2CH2O, 2ꢁH50), 3.79y (2ꢁ6H, s, OCH3 ), 4.27y,
y
4.38y (1H, br s, H40), 4.43y, 4.54y (1H, m, H30) 5.46y, 5.47y
(1H, d, H5, J5–6¼8.08 Hz, J5–6¼8.08 Hz), 5.59 (1H, br s,
NH-amide), 5.83 (1H, br s, NH-amide), 6.15y, 6.18y (1H, d,
3.6. Melting temperatures
H10, J1 –2 ¼9.68 Hz, J1 –2 ¼9.40 Hz), 6.83–6.85 (4H, m,
ArH o-OCH3), 7.22–7.42 (9H, m, ArH), 7.59y, 7.67y (1H, d,
H6 J6–5¼8.12 Hz, J6–5¼8.08 Hz),w10 (1H, br s, NH-uracil).
13P NMR d (161 MHz, CDCl3) 150.36, 151.36. Found HRMS
m/z [M+Na+], 810.3257, C41H50N5O9PNarequires810.3244.
0
0
0
0
Oligomers were dissolved in 0.1 M NaCl, 0.02 M potassium
phosphate, pH 7.5, 0.1 mM EDTA. The concentration was
determined by measuring the absorbance in MilliQ water
at 260 nm at 80 ꢀC and assuming the sugar modified nucleo-
side analogues to have the same extinction coefficients in the
denatured state as 20-deoxyuridine (3¼9660). The concen-
tration in all experiments was 4 mM for each strand unless
otherwise stated. Melting curves were determined with a
Varian Cary 300 BIO spectrophotometer. Cuvettes were
maintained at constant temperature by means of water circu-
lation through the cuvette holder. The temperature of the so-
lution was measured with a thermistor directly immersed in
the cuvette. Temperature control and data acquisition were
done automatically with an IBM-compatible computer using
Cary WinUV thermal application software. Following a
quick heating and cooling cycle to allow proper annealing
of both strands, the samples were heated at a rate of
0.2 ꢀC/min starting at 5 ꢀC up to 80 ꢀC and cooling again
at the same speed. Melting temperatures were determined
by plotting the first derivative of the absorbance versus tem-
perature curve and are the average of two runs. Up and down
curves in general showed identical Tm values.
3.4.4. 20-Deoxy-20-a-C-20-(2-amidoethyl)-50-O-dimethoxy-
trityluridine-[(2-cyanoethyl)-N,N-diisopropyl]phosphor-
amidite (25). 44%; 1H NMR d (400 MHz, CDCl3) 1.11–1.27
(12H, m, 4ꢁCH3), 1.76–1.83(1H, m, H60), 2.01–2.10(1H, m,
H600), 2.34–2.64 (5H, m, CH2CN, H20, H70, H700), 3.39–3.88
(6H, m, 2ꢁMe2CHN, NCCH2CH2O, 2ꢁH50), 3.80y (2ꢁ6H,
s, OCH3 ), 4.20y, 4.22y (1H, m, H40), 4.57–4.62y (2ꢁ1H, m,
y
H30 ) 5.33y, 5.34y (1H, d, H5, J5–6¼8.12 Hz, J5–6
¼
y
8.28 Hz), 5.46 (1H, br s, NH-amide), 5.90 (1H, br s,
0
NH-amide), 6.02y, 6.03y (1H, d, H10, J1 –2 ¼7.28 Hz,
0
0
0
J1 –2 ¼7.00 Hz), 6.83–6.86 (4H, m, ArH o-OCH3), 7.26–
7.40 (9H, m, ArH), 7.79y, 7.86y (1H, d, H6, J6–5¼8.24 Hz,
J6–5¼8.24 Hz), 9.25 (1H, br s, NH-uracil). 31P NMR
d (161 MHz, CDCl3) 149.73, 152.05. Found HRMS m/z
[M+Na+], 824.3405, C42H52N5O9PNa requires 824.3400.
3.5. Syntheses and purification of oligonucleotides
(26–29)
Acknowledgements
Oligonucleotides were prepared using an expedite automated
solid-phase synthesiser on a 1 mmol scale following standard
protocols. The exocyclic amine groups of adenosine and cy-
tidine were protected with benzoyl groups, while that of gua-
nine was protected with an isobutryl group. Dimethoxytrityl
groups were removed with 5% dichloroacetic acid. The
protected oligonucleotides were cleaved from the support
and protecting groups removed using aqueous ammonia,
55 ꢀC for 16 h. Final dimethoxytrityl groups were removed
following purification by HPLC, using 80% acetic acid/H2O.
We would like to thank the EPSRC for studentship to L.B.
We are also grateful to Allan Mills (Liverpool) and Dr.
Paul Leonard (Liverpool) for assistance with NMR spectros-
copy. We would also like to thank Guy Schepers (Leuven)
for assistance with oligo purification and Tm measurements
and Jef Rozenski (Leuven) for mass spectrometry.
Supplementary data
Further purification was done on a Mono QÒ-HR 5/5 anion
exchange column with the following gradient system
Supplementary data associated with this article can be found
in the online version, at doi:10.1016/j.tet.2006.11.024.

L. Brennan et al. / Tetrahedron 63 (2007) 577–585
585
14. Saneyoshi, H.; Seio, K.; Sekine, M. J. Org. Chem. 2005, 70,
10453.
References and notes
15. Sosnovsky, G.; Krogh, J.; Umhoefer, S. G. Communications
1979, 722.
1. Jansen, B.; Zangemeister-Wittke, U. Lancet Oncol. 2002,
3, 672.
16. Dorman, M. A.; Noble, S. A.; McBride, L. J. C.; Caruthers,
M. H. Tetrahedron 1984, 40, 95.
17. Reese, C. B. Tetrahedron 2002, 58, 8893.
18. Nurminen, E. J.; Mattinen, J. K.; Lonnberg, H. J. Chem. Soc.,
Perkin Trans. 2 2001, 2159.
19. De Voe, H.; Tinoco, I. J. Mol. Biol. 1962, 4, 500.
20. Cummins, L. L.; Owens, S. R.; Risen, L. M.; Lesnik, E. A.;
Freier, S. M.; McGee, D.; Guinosso, C. J.; Cook, P. D.
Nucleic Acids Res. 1995, 23, 2019.
21. Rich, A. Nat. Struct. Biol. 2003, 10, 247.
22. Wu, T.-F.; Froeyen, M.; Schepers, G.; Mullens, K.; Rozenski,
J.; Busson, R.; Van Aerschot, A.; Herdewijn, P. Org. Lett.
2004, 6, 51.
23. Ceulemans, G.; Van Aerschot, A.; Rozenski, J.; Herdewijn, P.
Tetrahedron 1997, 53, 14957.
24. Egli, M.; Minasov, G.; Tereshko, V.; Pallan, P. S.; Teplova, M.;
Inamati, G. B.; Lesnik, E. A.; Owens, S. R.; Ross, B. S.;
Prakash, T. P.; Manoharan, M. Biochemistry 2005, 44, 9045.
25. Lui, K.-T. H.; Hu, C.-J. Synthesis 1988, 7, 715.
2. Ichikawa, E.; Kato, K. Curr. Med. Chem. 2001, 8, 385.
3. Manoharan, M. Curr. Opin. Chem. Biol. 2004, 8, 570.
4. Chamberlin, M.; Berg, P. J. Mol. Biol. 1964, 8, 297.
5. Wang, S.; Kool, E. T. Biochemistry 1995, 34, 4125.
6. Crooke, S. T.; Lemondis, K. M.; Neilson, L.; Griffey, R.;
Lesnik, E. A.; Monia, B. P. Biochem. J. 1995, 312, 599.
7. Obika, S.; Nanbu, D.; Hari, Y.; Morio, K.; Ishida, T.; In, Y.;
Imanishi, T. Tetrahedron Lett. 1997, 38, 8735.
8. Singh, S. K.; Neilson, P.; Koshkin, A. A.; Wengel, J. Chem.
Commun. 1998, 455.
9. Freier, S. M.; Altmann, K. H. Nucleic Acids Res. 1997, 25,
4429.
10. Federoff, O.; Salazar, M.; Reid, B. R. J. Mol. Biol. 1993,
233, 509.
11. Gyi, J.; Lane, A. N.; Conn, G. L.; Brown, T. Biochemistry 1998,
37, 73.
12. Lawrence, A. J.; Pavey, J. B. J.; Cosstick, R.; O’Neil, I. A.
J. Org. Chem. 1996, 61, 9213.
13. Grotli, M.; Beijer, B.; Sproat, B. Tetrahedron 1999, 55, 4299.