Synthesis of 3-deoxy-3-carboxymethylnucleosides, precursors of oligonucleotides with an amide internucleoside bond

Article abstract of DOI:10.1134/S1068162009010099

An improved method for the synthesis of 3-deoxy-3-carboxymethyl nucleosides was suggested. Oxidation of 5-O-benzoyl-1,2-O-isopropylidene-α-D- xylofuranose resulted in the 3-keto derivative, which was treated with triethylphosphonoacetate in the presence o

Full text of DOI:10.1134/S1068162009010099

ISSN 1068-1620, Russian Journal of Bioorganic Chemistry, 2009, Vol. 35, No. 1, pp. 68–74. © Pleiades Publishing, Ltd., 2009.
Original Russian Text © S.V. Kochetkova, A.M. Varizhuk, N.A. Kolganova, E.N. Timofeev, V.L. Florent’ev, 2009, published in Bioorganicheskaya Khimiya, 2009, Vol. 35, No. 1,
pp. 76–83.
Synthesis of 3'-Deoxy-3'-Carboxymethylnucleosides, Precursors
of Oligonucleotides with an Amide Internucleoside Bond
S. V. Kochetkova, A. M. Varizhuk, N. A. Kolganova, E. N. Timofeev, and V. L. Florent’ev1
Engelhardt Institute of Molecular Biology, Russian Academy of Sciences, ul. Vavilova 32, Moscow, 119991 Russia
Received February 19, 2008; in final form, April 2, 2008
Abstract—An improved method for the synthesis of 3-deoxy-3-carboxymethyl nucleosides was suggested.
Oxidation of 5-O-benzoyl-1,2-O-isopropylidene-α-D-xylofuranose resulted in the 3-keto derivative, which was
treated with triethylphosphonoacetate in the presence of sodium hydride to obtain the 3-deoxy-3-ethoxycarbo-
nylmethylene derivative. Hydrogenation of the unsaturated compound proceeded strictly stereospecifically and
gave the product with the ribo-configuration. Acetolysis of the resulting compound with AcOH–Ac2O–
CH3SO3H led to 1,2-di-O-acetyl-5-O-benzoyl-3-deoxy-3-ethoxycarbonylmethyl-D-ribofuranose, whose inter-
action with persilylated nucleic bases gave 3-deoxy-3-ethoxycarbonylmethylnucleosides in a total yield of
42−49% from the starting compound.
Key words: branched carbohydrates, nucleoside analogues, synthesis
DOI: 10.1134/S1068162009010099
INTRODUCTION
with complementary natural oligonucleotides, than the
corresponding natural nucleotides (∆Mp = 0.5–0.9°C
for modification depending on the sequence [8, 9]).
Therapy with synthetic oligonucleotide analogues
(antisense therapy) is one of promising methods of
treatment of virus and oncological diseases [1–4].2 The
idea of the method consists in the suppression of gene
translation due to the specific binding of a synthetic oli-
gonucleotide with the corresponding mRNA site. The
high specificity of hybridization allows the selective
suppression of expression of the genes involved in
pathogenesis of disease. An antisense oligonucleotide
should however possess specific characteristics. It
should be stable to the action of endonucleases and at
the same time specifically and strongly contact with the
target mRNA sequence. Furthermore, it should pene-
trate easily through the cell membrane, possess low
toxicity and, show no side effects. For the last 15 years
a large number of oligonucleotide analogues modified
at the nucleic base, the carbohydrate residue, and at the
internucleotide bond (see [5–7]) have been synthesized.
Currently the oligonucleotide analogues with the amide
internucleoside bond attract the intent attention of
researchers. Such analogues initially possess at least
HO
Base
O
O
Base
O
X
(II)
X
O
O
OH + H2N
Base
O
NH
Base
O
OH X
a: X = H; b: X = OH
O
X
(I)
Scheme 1. Retrosynthetic analysis of oligonucleotides
amide internucleotide bond.
As seen from Scheme 1, 3'-deoxy-3'-carboxymethyl
nucleosides (II) are the key compounds for the synthe-
two characteristics of antisense oligonucleotides. They sis of such oligonucleotide analogues. 2',3'-Dideoxy-3'-
carboxymethyl nucleosides (IIa) were obtained
15 years ago [8, 10, 11]. However, ribonucleoside ana-
logues (IIb), which are the precursors of modified oli-
goribonucleotides, are of essentially greater interest.
Just the synthetic oligoribonucleotide analogues are
recently considered as promising regulators of biologi-
cal processes [12, 13].
are stable to endonucleases; the absence of the negative
charge promotes their easier penetration into cells. The
analogues with the C3'–CH2–CO–NH–C5' bond substi-
tuted for the phosphodiester bond (I) (Scheme 1) are of
especial interest, as they form the more stable duplexes
1
Corresponding author; phone: +7 (499) 135-2182, +7 (499) 135-
6591; fax: +7 (499) 135-1405; e-mail: flor@imb.ac.ru
2
Two methods for the synthesis of analogues (IIb)
from 2',5'-di-O-protected ribonucleosides [14, 15] have
Abbreviations: BSAA, N,O-bis(trimethylsilyl)acetamide; Tf, trif-
luoromethanesulfonyl.
68

SYNTHESIS OF 3'-DEOXY-3'-CARBOXYMETHYLNUCLEOSIDES
69
been published. The methods aforementioned have a [17] or ((IIb): B = Ura, Thy, or Ade) in 22–26% yield
number of disadvantages. In particular, this is the (9 steps) [18] calculated for 5-O-protected 1,2-O-iso-
necessity of selective protection of nucleoside 2'- and propylidene-α-D-xylofuranose.
5'-hydroxyl groups, but the main disadvantage is that
We herein report the improved method for the syn-
the methods are not general and do not allow the obtain-
thesis of nucleoside analogues (IIb: B = Ura, Thy, Cyt,
ing of all analogues with the natural bases, to say noth-
Ade, Gua) based on the approach afore-mentioned. The
ing of analogues with modified bases. The method of
method allows the obtaining of nucleoside analogues in
synthesis starting from the available compounds, which
five steps in 42–49% yield starting from 5-O-benzoyl-
uses rather simple and reproducible techniques and
1,2-O-isopropylidene-α-D-xylofuranose.
allows the obtaining of a nucleoside analogues both
with natural and modified bases, is preferred.
RESULTS AND DISCUSSION
Asymmetric synthesis starting from the achiral pre-
cursors of 1,2-di-O-acetyl-3-deoxy-3-tert-butyldiphe-
5-O-Benzoyl-1,2-O-isopropylidene-α-D-xylofuranose
nylsilyloxyethyl-D-ribose, which was then converted to
(III) [20, 21] was the starting compound in the synthe-
nucleoside analogues by the standard technique, was
suggested in [16]. After removal of the silyl protective
sis (Scheme 2). In [17, 18] compound (III) was oxi-
dized with CrO3–pyridine–Ac2O [22]. Though this
group, the hydroxyethyl group was oxidized to the car-
method results keto sugars in high yields, it is laborious
boxymethyl moiety. This method included 9–10 steps;
enough and rather sensitive to the ratio of reagents and
the yields of nucleoside analogues [(IIb): B = Ura, Cyt,
temperature. Oxidation of (III) with ruthenium tetrox-
Ade, and Gua] were 15–19% from the starting achiral
ide is described in CCl4 [23]. This reaction however
compounds.
requires very large volumes of the solvent and the yield
of ketone (IV) reaches only 55%. Oxidation with
sodium periodate in the presence of catalytic amount of
ruthenium dioxide in the two-phase chloroform–water
system gives very good results [24]. The attempt to use
this method toward xylose derivative (III) failed. It had
been previously shown [25] that sugar derivatives
hardly soluble in water can be oxidized in the system
under consideration in the presence of the interphase
transfer catalyst. Really, the addition of the catalytic
amount of triethylbenzylammonium chloride to the
reaction mixture resulted in the formation of ketone
The method [17, 18] based on the approach to the
synthesis of branched sugars suggested earlier [19] has
been reported. This method included oxidation of 5-O-
protected
1,2-O-isopropylidene-α-D-xylofuranose,
treating the resulting 3-ulose with [(ethoxycarbo-
nyl)methylene]triphenylphosphorane, hydrogenation
of the adduct, and opening of the isopropylidene cycle
by treating with either HCl in methanol with the subse-
quent acetylation [17], or AcOH–Ac2O–H2SO4 [18].
The resulting methyl-2-O-acetylriboside or 1,2-di-O-
acetyl derivative were used for the synthesis of nucleo-
side analogues ((IIb): B = Thy) in 16% yield (5 steps) (IV) in 83% yield.
5
BzO
BzO
BzO
H
O
O
O
1
4
OH
‡
b
3
2
O
O
O
CH3
CH3
CH3
CH3
CH3
CH3
O
6
O
O
O
O
O
(III)
(IV)
CH3
B
(V)
H5'a
BzO
O
BzO
O
O
H5'b
O
BzO
O
OAc
d
c
e
O
CH3
CH3
(VI)
OAc
CH3
(VII)
a – Thy; b – Ura; c – Cyt; d – Ade; e – N-AcGua
OAc
O
O
H6'b
H6'a
CH3
O
O
CH3
O
(VIII)
Scheme 2. Reagents and conditions: a, NaIO , RuO , Et BnNCl, chloroform/water; b, (EtO) P(O)CH COOEt, NaH, THF; c, H ,
4
2
3
2
2
2
Pd(OH) /C, ethanol; d, AcOH, Ac O, MeSO OH; e, nucleic bases, BSA, TMSOTf, acetonitrile
2
2
2
RUSSIAN JOURNAL OF BIOORGANIC CHEMISTRY Vol. 35 No. 1 2009

70
KOCHETKOVA et al.
H6b
H6a
H3
4.1
4.2
4.3
4.4
4.5
4.6
4.7
4.8
4.9
2.8
2.7
2.6
2.5
2.4
2.3
f2, ppm
A fragment of the NOESY spectrum of 5-O-benzoyl-3-deoxy-3-ethoxycarbonylmethyl-1,2-O-isopropylidene-α-D-ribofuranose
(VI), which displays the interaction of protons H2, H5a, H5b, and H4 with protons H6a, H6b, and H3. The signal of proton H4
overlaps with the signal of the ëç protons of the ethoxy group.
2
Ketone (IV) was converted to unsaturated com- were calculated for the 10 most stable conformers
pound (V) by the Horner–Wadsworth–Emmons reac- (Table 1).
tion [26]. The resulting mixture of E- and Z-isomers
The computation showed that the bicyclic system is
1
(7 : 1, according to H NMR spectroscopy data) was
conformationally rigid and the distances H2–H3 and
H3–H4 are practically independent from the conforma-
tion of side chains, H2 and H3 being in the nearest posi-
tion in the ribo-configuration and H3 and H4 being in
the nearest position in the xylo-configuration. As seen
hydrogenated under atmospheric pressure to result in
the only compound (VI) in the almost quantitative
yield, whose ribo-configuration was corroborated by
the NOESY-spectrum (see the figure).
The possible conformers for stereoisomers with the from the figure, just the H2–H3 cross-peak is the most
ribo-and the xylo-configuration were computed. The intensive. Furthermore, H4–H6a and H4−H6b cross
average values (1/r)6, where r is the interproton distance peaks should be observed in the case of the ribo-config-
RUSSIAN JOURNAL OF BIOORGANIC CHEMISTRY Vol. 35 No. 1 2009

SYNTHESIS OF 3'-DEOXY-3'-CARBOXYMETHYLNUCLEOSIDES
71
Table 1. The values (1/r)6 for 5-O-benzoyl-3-deoxy-3-ethoxycarbonylmethyl-1,2-O-isopropylidene-α-D-ribo- and -α-D-
xylofuranose averaged for the most stable conformers (the values of (1/r)6 were normalized to unity for ease of comparison)
The values of (1/r)6 for proton pairs
Config-
uration
H2–H3
H3–H4
H4–H6a*
H4–H6b* H6a*–H5a* H6a*–H5b* H6b*–H5a* H6b*–H5b*
ribo
xylo
1.00
0.39
0.22
1.00
0.23
0.05
0.56
0.05
0.05
0.27
0.05
0.20
0.19
0.25
0.11
0.21
Note: * 5a, pro-S-proton of the 5-CH group; 6a, pro-R-proton of the 6-CH -group (the spatial arrangement of these protons is shown in Scheme 2).
2
2
uration, and they should not be observed in the xylo- H3 and two signals H6a and H6b being doublets of
configuration that is in a good agreement with the doublets. The large heminal coupling constant (J =
experimental spectrum. Absence of H6a–H5a and 16.61 Hz) is a specialty of the 6-CH2 group. This spe-
H6a−H5b cross peaks in the NOESY-spectrum is an cialty retains in all products obtained from (VI). The
additional confirmation of the ribo-configuration of the spectrum of diacetate (VII) contains no signal of CH3
compound obtained.
fragments of the isopropylidene group and two singlets
of acetyl groups.
The spectra of the nucleosides (given in the Experi-
mental) retain the specialties of the spectrum of diace-
tate and also contain the signals of the nucleic base.
Thus, the proposed method allows us to obtain the
nucleoside analogues in five steps in a yield of 42–49%
starting from 5-O-benzoyl-1,2-O-isopropylidene-α-D-
xylofuranose.
The removal of the isopropylidene group and the
obtaining of 2-O-acetyl derivative is the key stage of the
synthesis. In the study [17] isopropylidene derivative
(VI) was treated with 1% HCl in methanol and then
acetylated without isolation. However, methyl-5-O-
benzoyl-3-deoxy-3-methoxycarbonylmethyl-2-O-acetyl-
D-ribofuranoside was obtained only in 41% yield.
Peterson et al. [18] demonstrated that acetolysis of
5-O-acetyl-3-deoxy-3-ethoxycarbonylmethyl-1,2-O-iso-
propylidene-α-D-ribofuranose with AcOH–Ac2O–H2SO4
proceeds with a low yield (25%). The authors found that
the substitution of the 5-O-methanesulfonyl group for
5-O-acetyl group essentially raises (up to 70%) the yield
of acetolysis. Therefore, they introduced two additional
steps in the synthetic scheme, namely, the substitution of
the 5-O-methanesulfonyl group for 5-O-acetyl group and
the reverse substitution of the 5-O-acetyl group for the
5-O-methanesulfonyl group after acetolysis. The total
yield from the three steps was 51%.
EXPERIMENTAL
Benzene, chloroform, acetonitrile, ethyl acetate, tet-
rahydrofuran, and acetic acid (all reagent grade) were
from Khimmed (Russia); acetic anhydride (reagent grade)
was from Reakhim (Russia); cytosine, thymine, adenine,
and guanine were from Fluka (Switzerland); trieth-
ylphosphonoacetate, N,O-bis(trimethylsilyl)acetamide,
trimethylsilyltrifluoromethanesulfonate, sodium hydride,
ruthenium dioxide, and sodium periodate were from Ald-
rich (United States). Benzene and acetonitrile were dried
by distillation from phosphorus pentoxide; tetrahydrofu-
ran was dried by distillation from lithium aluminum
hydride. 2-N-Acetylguanine was obtained by the method
[28].
NMR spectra (δ, ppm, J, Hz) were registered in
CDCl3 (when it is not especially specified) on a Bruker
AMXIII-400 spectrometer at a working frequency of
400 MHz.
Conformational computation was carried out by the
Monte−Carlo method (Conformational Search option)
using HyperChem 8.04 (HyperCube, Canada) program.
We succeeded to increase the yield of diacetate
(VII) up to 82% by the substitution of methanesulfonic
acid for sulfuric acid on and by the choice of reagent
ratio. The resulting diacetate (VII) was used for the
synthesis of nucleoside analogues (VIIIa)–(VIIIe) by
to some extend modified standard procedure [27].
The structure of all compounds obtained was con-
firmed with 1H NMR spectra (Table 2). As can be seen,
the spectrum of ketone (IV) has no signal of H3, and
the signal of H1 is shifted downfield for 0.219 ppm as
compared to its position in the spectrum of the starting
compound (III) that is due to the effect of the carbonyl
group anisotropy. Furthermore, the distant (through
four bonds) coupling constant (J = 1.2 Hz) between H2
and H4 pointing to the sp2-hybridization of the C3 atom
is observed. The signal of H6 being a triplet due to the
distant (allyl) coupling constant with H2 and H4
appears at 6.149 ppm in the spectrum of compound (V).
The signal of H2 appears as a doublet of triplets
(because of interaction with H4 and H6). Hydrogena-
tion of the double bond in compound (VI) results in the
Column chromatography was carried out on Silica
gel 60 (Merck, Germany). TLC was carried out on Sil-
ica gel 60 F254 (Merck, Germany) precoated palates in
(A) 1 : 19 methanol–benzene, (B) 1 : 1 chloroform–
hexane containing 3% isopropanol, (C) 1 : 19 isopro-
panol–hexane, (D) 1 : 19 ethanol–chloroform, and (E)
1 : 9 ethanol–chloroform.
5-é-Benzoyl-1,2-Oisopropylidene-a-D-xylofura-
appearance of expected upfield (2.390 ppm) signal of nose (III) was obtained from D-xylose in two steps
RUSSIAN JOURNAL OF BIOORGANIC CHEMISTRY Vol. 35 No. 1 2009

72
Table 2. The 1H NMR data for (III)–(VII) (solutions in CDCl3)
Chemical shifts, ppm (J, Hz)
KOCHETKOVA et al.
Protons
(III)
(IV)*
(V)*
(VI)
(VII)
6.116 br. s
H1
5.947 d (J1.2 3.53)
4.582 d (J1.2 3.53)
6.166 d (J1,2 4.47)
5.958 d (J1,2 4.14)
5.865 d (J1.2 3.97)
4.681 dd (J1,2 4.47,
4J2,4 1.2)
5.653 dt (J1,2 4.14,
4J2,4(6) 1.52)
4.819 pseudo t
(J1.2 3.97, J2,3 4.51)
5.341 d (J2,3 4.73)
H2
H3
4.174 d (J3,4 4.75)
2.390 m
4.112 m
2.967 m
4.295 m
H4
4.343–4.416 m
4.864 m
5.183 m
4.789 dd (J4,5a 9.29, 4.536 dd (J4,5a 2.92, 4.534 dd (J4,5a 3.11, 4.558 dd (J4,5a 2.70, 4.603 dd (J4,5a 2.16,
H5a
2J5a,5b 12.84)
2J5a,5b 12.19)
2J5a,5b 12.04)
2J5a,5b 12.26)
2J5a,5b 12.11)
4.343–4.416 m
4.434 dd (J4,5b 4.35, 4.467 dd (J4,5b 4.65, 4.336 dd (J4,5b 5.05, 4.324 dd (J4,5b 4.89,
H5b
2J5a,5b 12.19)
2J5a,5b 12.04)
6.149 t (4J6.2(4) 1.81) 2.745 dd (J3,6a 9.89, 2.595 dd (J3,6a 8.81,
2J6a,6b 16.97) 2J6a,6b 16.44)
2J5a,5b 12.26)
2J5a,5b 12.11)
H6a
H6b
2.484 dd (J3,6b 4.38, 2.511 dd (J3,6b 6.00,
2J6a,6b 16.97)
1.507 s
2J6a,6b 16.44)
CH3iPr
1.494 s
1.314 s
1.419 s
1.363 s
1.384 s
1.331 s
CH3iPr
1.321 s
OCH2CH3
OCH2CH3
1-OAc
4.148 quart (J 7.09) 4.135 quart (J 7.13) 4.134 quart (J 7.14)
1.216 t (J 7.09)
1.246 t (J 7.13)
1.237 t (J 7.14)
1.950 s
2-OAc
2.107 s
o-H Bz
8.043 m
7.445 m
7.581 m
7.932 m
7.532 m
7.676 m
7.925 m
7.534 m
7.668 m
8.040 m
7.432 m
7.559 m
8.057 m
7.433 m
7.564 m
m-H Bz
n-H Bz
* Registered in DMSO-d .
6
according to the methods described in [20, 21] in a total hard oil; Rf 0.36 (Ä). For the 1ç NMR spectral data see
1
Table 2.
yield of 67%. H NMR spectral data are given in
Table 2.
5-é-Benzoyl-3-deoxy-3-ethoxycarbonylmetylene-
1,2-O-isopropylidene-a-D-ribofuranose (V). A 60%
suspension of sodium hydride in mineral oil (0.7 g,
17.6 mmol) and anhydrous THF (10 ml) were placed in
a flask supplied by a condenser with a calcium chloride
tube, a drop funnel and a thermometer. A solution of tri-
ethylphosphonoacetate (3.95 g, 17.6 mmol) in anhy-
drous THF (10 ml) was added dropwise to the mixture,
so that the temperature did not exceed 5°ë. After the
addition was completed, the mixture was stirred for
additional 20 min at 0–5°ë. Then a solution of ketone
(IV) (4.7 g, 16.1 mmol) in anhydrous THF (10 ml) was
added (the temperature should not exceed 5°ë). Cool-
ing was removed; the mixture was stirred for1 h at room
temperature and for 30 min at 50°ë. The mixture was
diluted with water (50 ml) and extracted with ethyl ace-
tate (3 × 50 ml). The combined organic extract was
washed with water and dried with anhydrous Na2SO4.
The solvent was removed; the residue chromato-
graphed on a Silica gel column (4 × 20 cm) in 1 : 1 chlo-
5-é-Benzoyl-1,2-O-isopropylidene-a-D-erythro-
pentafuranos-3-ulose (IV). Water (50 ml), ruthenium
dioxide (0.1 g), NaIO4, (12.8 g, 60 mmol), K2CO3
(1.0 g, 7.2 mmol), and benzyltriethylammonium chlo-
ride (0.136 g, 0.6 mmol) were added to a solution of
compound (III) (5.88 g, 20 mmol) in chloroform
(50 ml) free of alcohol at stirring. The mixture was
stirred for 24 h at a room temperature (TLC control in
system A). Then isopropanol (10 ml) was added; the
mixture was stirred for 10 min and then filtered through
Celite, which washed with chloroform (3 × 10 ml) on
the filter. The organic layer was separated; the aqueous
layer was extracted with chloroform (2 × 30 ml). The
combined chloroform extract was washed with satu-
rated aqueous NaHCO3 (50 ml) and water and dried
with anhydrous Na2SO4; the solvent was evaporated in
a vacuum. The residue was re-evaporated with 1 : 1
anhydrous benzene–anhydrous acetonitrile (4 × 10 ml)
and dried in a vacuum to give 4.86 g (83%) of (IV) as roform–hexane containing isopropanol (a linear gradi-
RUSSIAN JOURNAL OF BIOORGANIC CHEMISTRY Vol. 35 No. 1 2009

SYNTHESIS OF 3'-DEOXY-3'-CARBOXYMETHYLNUCLEOSIDES
73
1
ent from 1 to 5% isopropanol). The fractions containing (81%) as firm foam, Rf = 0.36 (D). H NMR: 8.725
the target product [the mixture of two isomers in a ratio (1 H, s, H3), 8.042 (2 H, m, Ó-H Bz), 7.588 (1 H, m,
of 1 : 7, Rf = 0.37 and 0.46 (B)], were combined; the sol- p-H Bz), 7.454 (2 H, m, m-H Bz), 7.116 (1 H, m, J =
vent was removed in a vacuum to give 5.8 g (91%) of 1.17, H6), 5.808 (1 H, d, J1',2' = 2.60, H1'), 5.514 (1 H,
(V) as hard oil. The 1H NMR spectral data are given in dd, J1',2' = 2.60, J2',3' = 6.85, H2'), 4.717 (1 H, dd, J4',5'a
=
Table 2.
2.41, 2J5'a,5'b = 12.62, H5'a), 4.473 (1 H, dd, J4',5'b = 4.07,
2J5'a,5'b = 12.62, H5'b), 4.245 (1 H, m, H4'), 4.117 (2 H,
quart, J = 7.14, OCH2CH3), 2.994 (1 H, m, H3'), 2.595
(1 H, dd, J3',6'a = 8.81, 2J6'a,6'b = 16.44, H6'a), 2.511 (1 H,
5-é-Benzoyl-3-deoxy-3-ethoxycarbonylmethyl-1,2-
O-isopropylidene-a-D-ribofuranose (VI). Pd(OH)2/C
(0.57 g) was added to a solution of compound (V) (5.7 g,
15.7 mmol) in ethanol (30 ml); the mixture was hydroge-
nated under atmospheric pressure and room temperature for
3 days. The mixture filtered through Celite, which was then
washed on the filter with ethanol (3 × 10 ml). The solvent
was removed in a vacuum to give 5.6 g (98%) of (VI) as
hard oil, Rf = 0.46 (B). The 1H NMR spectral data are given
in Table 2.
2
dd, J3',6'b = 6.00, J6'a,6'b = 16.44, H6'b), 2.114 (3 H, s,
2'-OCOCH3), 1.658 (3 H, d, J = 1.31, 5-CH3), 1.237
(3 H, t, J = 7.14, OCH2CH3).
2'-O-Acetyl-5'-é-benzoyl-3'-deoxy-3'-ethoxycar-
bonylmethyluridine (VIIIb) was obtained from uracil
(0.28 g, 2.5 mmol) and BSAA (1.12 g, 1.3 ml,
5.5 mmol). Elution with chloroform–ethanol (a linear
gradient from 2 to 6% ethanol). Yield 0.67 g (73%) as
firm foam, Rf = 0.39 (D); 1H NMR: 8.441 (1 H, s, H3),
8.025 (2 H, m, Ó-H Bz), 7.603 (1 H, m, p-H Bz), 7.467
(2 H, m, m-H Bz), 7.430 (1 H, d, J5.6 = 8.24, H6), 5.784
(1 H, d, J1',2' = 1.67, H1'), 5.527 (1 H, dd, J1',2' = 1.67,
J2',3' = 5.40, H2'), 5.522 (1 H, d, J5.6 = 8.24, H5), 4.703
(1 H, dd, J4',5'a = 2.37, 2J5'a,5'b = 12.67, H5'a), 4.517 (1 H,
1,2-Di-O-acetyl-5-é-benzoyl-3-deoxy-3-ethoxy-
carbonylmethyl-D-ribofuranose (VII). A mixture of
glacial acetic acid (60 ml) and acetic anhydride (15 ml)
was stirred for 30 min at room temperature. Then iso-
propylidene derivative (VI) (5.5 g, 15.1 mmol) was
added. The resulting solution was cooled with ice;
methanesulfonic acid (6 ml) was slowly flowed at vig-
orous stirring. Cooling was removed; the reaction mix-
ture was stirred at room to temperature for 24 h, poured
slowly to the mixture of water and ice (300 ml), and
stirred for 45 min. The mixture was extracted with chlo-
roform (3 × 50 ml); the extracts were washed with satu-
rated aqueous NaHCO3 (3 × 50 ml) and water, dried
with anhydrous Na2SO4, and evaporated. The residue
was chromatographed on a Silica gel column (4 ×
20 cm) in 1 : 1 chloroform–hexane containing isopro-
panol (a linear gradient from 1 to 5% isopropanol). The
fractions containing the target product [Rf = 0.36 (B)]
were combined; the solvent was evaporated to give 5.1 g
(83%) of (VII) as hard oil, Rf = 0.36 (B). The 1H NMR
spectral data are given in Table 2.
2
dd, J4',5'b= 3.99, J5'a,5'b = 12.67, H5'b), 4.269 (1 H, m,
H4'), 4.120 (2 H, quart, J = 7.13, OCH2CH32), 2.929
(1 H, m, H3'), 2.582 (1 H, dd, J3',6'a = 8.21, J6'a,6'b
=
16.61, H6'a), 2.496 (1 H, dd, J3',6'b = 6.23, 2J6'a,6'b = 16.61,
H6'b), 2.124 (3 H, s, 2'-OCOCH3), 1.225 (3 H, t, J 7.13,
OCH2CH3).
2'-O-Acetyl-5'-é-benzoyl-3'-deoxy-3'-ethoxycar-
bonylmethylcytidine (VIIIc) was obtained from
cytosine (0.28 g, 2.5 mmol) and BSAA (1.68 g, 2 ml,
8.3 mmol). Elution with chloroform–ethanol (a linear
gradient from 5 to 12% ethanol). Yield 0.77 g (84%) as
firm foam, Rf = 0.27 (E); 1H NMR: 8.025 (2 H, m, o-H
Bz), 7.595 (1 H, d, J5,6 = 7.41, H6), 7.586 (1 H, m, p-H
Bz), 7.465 (2 H, m, m-H Bz), 5.828 (1 H, d, J1',2' = 1.91,
H1'), 5.705 (1 H, d, J5,6 = 7.41, H5), 5.569 (1 H, dd, J1',2'
1.91, J2',3' 5.97 = 1.91, J2',3' = 5.97, H2'), 4.677 (1 H, dd,
J4',5'a = 2.46, 2J5'a,5'b = 12.61, H5'a), 4.525 (1 H, dd, J4',5'b
= 4.39, 2J5'a,5'b = 12.61, H5'b), 4.283 (1 H, m, H4'), 4.087
(2 H, quart, J = 7.13, OCH2CH3), 2.884 (1 H, m, H3'),
2.551 (1 H, dd, J3',6'a = 7.86, 2J6'a,6'b = 16.66, H6'a), 2.437
(1 H, dd, J3',6'b = 6.26, 2J6'a,6'b = 16.66, H6'b), 2.113 (3 H,
s, 2'-OCOCH3), 1.229 (3 H, t, J 7.13, OCH2CH3).
The general procedure for obtaining nucleoside
analogues (VIIIa)–(VIIIe). BSAA (1.1 mmol for each
NH group of the base) was added to a suspension of a
nucleic base (2.5 mmol) in anhydrous acetonitrile
(5 ml) at stirring; the mixture was heated at 50°ë up to
dissolution of the precipitate (0.5–1 h). A solution of
diacetate (VII) (0.82 g, 2 mmol) in anhydrous acetoni-
trile (4 ml) and then TMSOTf (0.56 g, 0.45 ml,
2.5 mmol) were added to the resulting solution. The
mixture was refluxed for 4 h, cooled, poured to cold sat-
urated aqueous NaHCO3 (50 ml) at stirring, stirred for
30 min, and extracted with chloroform (3 × 20 ml). The
extracts were washed with water (20 ml) and dried with
anhydrous Na2SO4. The solvent was removed in a vac-
uum; the residue was chromatographed on a Silica gel
column (3 × 20 cm).
2'-O-Acetyl-5'-é-benzoyl-3'-deoxy-3'-ethoxycar-
bonylmethyladenosine (VIIId) was obtained from
adenine (0.34 g, 2.5 mmol) and BSAA (1.68 g, 2 ml,
8.3 mmol). Elution with chloroform–ethanol (a linear
gradient from 1 to 6% ethanol). Yield 0.67 g (69%) as
firm foam, Rf = 0.34 (D); 1H NMR: 8.249 (1 H, s, H2),
7.949 (1 H, s, H8), 7.938 (2 H, m, o-H Bz), 7.551 (1 H,
m, p-H Bz), 7.404 (2 H, m, m-H Bz), 5.998 (1 H, d,
2'-O-Acetyl-5'-é-benzoyl-3'-deoxy-3'-ethoxycar- J1',2' = 1.61, H1'), 5.923 (1 H, dd, J1',2' = 1.61, J2',3' = 5.92,
bonylmethylribosylthymine (VIIIa) was obtained H2'), 4.694 (1 H, dd, J4',5'a = 2.63, 2J5'a,5'b = 12.48, H5'a),
from thymine (0.32 g, 2.5 mmol) and BSAA (1.12 g, 4.694 (1 H, dd, J4',5'b = 4.74, 2J5'a,5'b = 12.48, H5'b), 4.381
1.3 ml, 5.5 mmol). Elution with chloroform–ethanol (a (1 H, m, H4'), 4.129 (2 H, quart, J 7.13, OCH2CH3),
linear gradient from 2 to 6% ethanol).Yield 0.77 g 3.529 (1 H, m, H3'), 2.679 (1 H, dd, J3',6'a = 8.58,
RUSSIAN JOURNAL OF BIOORGANIC CHEMISTRY Vol. 35 No. 1 2009

74
KOCHETKOVA et al.
2J6'a,6'b = 16.52, H6'a), 2.602 (1 H, dd, J3',6'b = 6.02,
2J6'a,6'b = 16.52, H6'b), 2.144 (3 H, s, 2'-OCOCH3), 1.226
(3 H, t, J 7.13, OCH2CH3).
11. De Mesmaeker, A., Lesueur, C., Bevierre, M.-O., Wald-
ner, A., Fritsch, V., and Wolf, R.M., Angew. Chem., Int.
Ed. Engl., 1996, vol. 35, pp. 1970–1974.
12. Manoharan, M., Curr. Opin. Chem. Biol., 2004, vol. 8,
2'-O-Acetyl-5'-é-benzoyl-3'-deoxy-3'-ethoxycar-
pp. 570–579.
bonylmethyl-2-N-acetylguanosine
(VIIIe)
was
obtained from N-2-acetylguanine (0.48 g, 2.5 mmol)
and BSAA (1.68 g, 2 ml, 8.3 mmol). Elution with chlo-
roform–ethanol (a linear gradient from 1 to 5% etha-
nol).Yield 0.78 g (72%) as firm foam, Rf = 0.22 (D); 1H
NMR: 7.838 (1 H, s, H8), 7.786 (2 H, m, o-H Bz), 7.555
(1 H, m, p-H Bz), 7.368 (2 H, m, m-H Bz), 5.930 (1 H,
13. Fichou, Y. and Férec, C., Trends Biotechnol., 2006,
vol. 24, pp. 563–570.
14. Peterson, M.A., Nilsson, B.L., Sarker, S., Doboszewski, B.,
Zhang, W., and Robins, M.J., J. Org. Chem., 1999, vol. 64,
pp. 8183–8192.
dd, J1',2' = 0.84, J2',3' = 5.28, H2'), 5.855 (1 H, d, J1',2'
=
15. Rozners, E., Katkevica, D., Bizdena, E., and Stromberg, R.,
0.84, H1'), 4.759 (1 H, dd, J4',5'a = 3.58, 2J5'a,5'b = 12.39,
H5'a), 4.595 (1 H, dd, J4',5'b = 3.23, 2J5'a,5'b = 12.39, H5'b),
4.315 (1 H, m, H4'), 4.124 (2 H, quart, J 7.13,
OCH2CH3), 3.841 (1 H, m, H3'), 2.581 (2 H, m, H6'a,
H6'b), 2.177 (3 H, s, 2-NCOCH3), 2.151 (3 H, s, 2'-
OCOCH3), 1.248 (3 H, t, J 7.13, OCH2CH3).
J. Am. Chem. Soc., 2003, vol. 125, pp. 12 125–12 136.
16. Rozners, E. and Liu, Y., J. Org. Chem., 2005, vol. 70,
pp. 9841–9848.
17. Lu, D.-M., Min, J.-M., and Zhang, L.-H., Carbohydr.
Res., 1999, vol. 317, pp. 193–197.
18. Robins, M.J., Doboszewski, B., Timoshchuk, V.A., and
Peterson, M.A., J. Org. Chem., 2000, vol. 65, pp. 2939–
2945.
ACKNOWLEDGMENTS
Work was supported by the Russian Foundation for
Basic Research (project no. 08-04-01078).
19. Rosenthal, A. and Nguyen, B., Tetrahedron Lett., 1967,
vol. 25, pp. 2393–2396.
20. Moravcova, J., Capkova, J., and Stanek, J., Carbohydr.
Res., 1994, vol. 263, pp. 61–66.
REFERENCES
21. Molina, J., Maslen, H.L., and Simons, C., Nucleosides
Nucleotides Nucleic Acids, 2001, vol. 20, pp. 981–983.
1. Dias, N. and Stein, C.A., Mol. Cancer. Ther., 2000,
vol. 1, pp. 347–355.
22. Garegg, P.J. and Samuelsson, B., Carbohydr. Res., 1978,
2. Taylor, M.F., Drug Discovery Today, 2001, vol. 6,
vol. 67, pp. 267–270.
pp. 97–101.
23. Nutt, R.F, Dickinson, M.J, Holly, F.W, and Walton, E, in
Synthetic Procedures in Nucleic Acid Chemistry, Zor-
bach, W.W.. and Tipson, R.S., Eds., New York: Inter-
science Publishers, John Wiley & Sons, 1968, pp. 54–
159.
3. Fichou, Y. and Férec, C., Trends Biotechnol., 2006,
vol. 24, pp. 563–570.
4. Leonetti, C. and Zupi, G., Curr. Pharm. Des., 2007,
vol. 13, pp. 463–470.
5. Micklefield, J., Curr. Med. Chem., 2001, vol. 8,
24. Baker, D.C., Horton, D., and Tindall, C.G., Carbohydr.
Res., 1972, vol. 24, pp. 192–197.
pp. 1157–1179.
6. Kurreck, J., Eur. J. Biochem., 2003, vol. 270, pp. 1628–
25. Morris, P.E. and Kiely, D.E., J. Org. Chem., 1987,
1644.
vol. 52, pp. 1149–1152.
7. Tomohisa, M. and Kazuo, S., Frontiers in Organic
Chemistry, 2006, vol. 1, pp. 163–194.
26. Wadsworth, W.S, in Organic Reaction, Dauben, W.G.,
Ed., Malabar, FL: Krieger Publishing Co., 1977, vol. 25,
pp. 73–244.
8. De Mesmaeker,A., Waldner,A., Lebreton, J., Hoffmann, P.,
Fritsch,V., Wolf, R.M., and Freier, S.M., Angew. Chem., Int.
Ed. Engl., 1994, vol. 33, pp. 226–229.
27. Vorbruggen, H, Lagoja, I.M, and Herdewijn, P, in Cur-
rent Protocols in Nucleic Acid Chemistry, Glick, G.D.
and Jones, R.A., Eds., New York: Wiley Interscience,
Chapter 1, Unit 1.13.
9. von Matt, P., Mesmaeker, D.A., Pieles, U., Zurcher, W.,
and Altmann, K.-H., Tetrahedron Lett., 1999, vol. 40,
pp. 2899–2902.
28. Hrebabecky, H and Farkas, J, in Nucleic Acid Chemistry,
Townsend, L.B.. and Tipson, S., Eds., New York: Inter-
science Publishers, John Wiley & Sons, 1978, Part I,
pp. 13–15.
10. De Mesmaeker, A., Lebreton, J., Waldner, A., Fritsch, V.,
Wolf, R.M., and Freier, S.M., Synlett., 1993, no. 10,
pp. 733–736.
RUSSIAN JOURNAL OF BIOORGANIC CHEMISTRY Vol. 35 No. 1 2009