Synthesis of 3′-C-substituted thymidine derivatives by free-radical techniques: scope and limitations

Article abstract of DOI:10.1016/j.carres.2006.11.015

The scope and limitations of radical-mediated 3′-C-substitution of pyrimidine nucleosides was evaluated with 5′-O-(tert-butyldimethylsilyl)thymidine or its tert-butyldiphenylsilyl analogue having thionoester or thionoamide groups at O-3′, including (methy

Full text of DOI:10.1016/j.carres.2006.11.015

Carbohydrate Research 342 (2007) 259–267
Note
Synthesis of 30-C-substituted thymidine derivatives
by free-radical techniques: scope and limitations
Derek Horton,* Kuangmin Chen, Zaesung No and Howard C. Lee
Department of Chemistry, American University, 4400 Massachusetts Avenue, Washington, DC 20016, USA
Received 1 September 2006; received in revised form 7 November 2006; accepted 11 November 2006
Available online 17 November 2006
Abstract—The scope and limitations of radical-mediated 30-C-substitution of pyrimidine nucleosides was evaluated with 50-O-(tert-
butyldimethylsilyl)thymidine or its tert-butyldiphenylsilyl analogue having thionoester or thionoamide groups at O-30, including
(methylthio)thiocarbonyl, (phenoxy)thiocarbonyl, (pentafluorophenoxy)thiocarbonyl, and (1-imidazolyl)thiocarbonyl. Their reac-
tion with acrylonitrile, methyl acrylate, and allyltributyltin under radical-generating conditions affords corresponding 30-C-alkylated
products, together with the product of simple deoxygenation at C-30. The conditions for optimizing the yield of 30-C-substituted
product are presented.
ꢀ 2006 Elsevier Ltd. All rights reserved.
Keywords: Free-radical; Thymidine; Nucleoside; Thionoamide; Thionoester; C-Alkylation
Nucleoside mimetics have held sustained interest over
many decades for their potential in the treatment of dis-
eases, as antitumor, antiviral, antibacterial, or antipara-
sitic agents.1 Our laboratory has explored synthetic
methodology targeted especially toward nucleoside ana-
logues modified in the sugar portion, and to structures
having a noncyclized glycon component.2 The antiviral
potential of such mimetics has stimulated much activity,
especially in the quest for compounds effective in treat-
ing acquired immunodeficiency syndrome (AIDS).3
Such agents include3 the reverse-transcriptase inhibitors
AZT (Zidovudine), 3TC (Lamivudine), ABC (Abaca-
vir), d4T, ddI, and ddC, which are all 20,30-dideoxy
nucleoside derivatives. The ability of the human immuno-
deficiency virus (HIV) to become resistant to a particu-
lar drug regimen emphasizes the need for newer agents.4
In this context we have evaluated synthetic methodology
directed toward C-30-substituted 20,30-dideoxy nucleo-
side derivatives.5 Examples of such compounds include
30-fluoro-30-deoxythymidine,6 a potent anti-HIV agent
in vitro, and C-30-alkenyl (and alkynyl)-30-
deoxythymidines.7
We sought general and high-yielding procedures for
C-30-alkylated 20,30-dideoxy nucleosides incorporating
a variety of alkyl chains. Possible synthetic approaches
to C-30-substituted nucleoside derivatives include proce-
dures based on the attack of a nucleophile on a pento-
furanos-3-ulose derivative, either on a precursor sugar
derivative or on a preformed 30-keto nucleoside. The lat-
ter approach often leads simply to the elimination of the
nucleoside base, and the former requires an additional
synthetic step for covalent attachment of the base. In re-
cent years, free-radical chemistry has been shown to be
an attractive alternative for synthesizing C-30-substi-
tuted 20,30-dideoxynucleoside analogues directly from a
preformed nucleoside, as documented notably in the
work of Chu et al.8 and others,9,10 who introduced a
30-C-allyl group in a 50-protected thymidine bearing a
thiono substituent at C-30. The scope and limitations
of this approach with respect to other carbon donors
remains to be explored.
Our studies have likewise employed thymidine as the
precursor, protected at O-50 by the tert-butyldimethyl-
silyl11 or tert-butyldiphenylsilyl12 groups, and applica-
tion of the general strategy of Keck and Yates for
allylation of a carbon radical with allyltributyltin13
and of Giese et al.14 for C-glycosylation of hexoses with
*
Corresponding author. Tel.: +1 202 885 1767; fax: +1 202 885
1095; e-mail: carbchm@american.edu
0008-6215/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2006.11.015

260
D. Horton et al. / Carbohydrate Research 342 (2007) 259–267
acrylonitrile and tributyltin hydride. Optimization of the
procedure involved examination of various radical pre-
cursors and the evaluation of different trapping agents
and experimental conditions.
crystalline in 25% yield, was the desired 30-alkylated
product, namely 50-O-tert-butyldimethylsilyl-30-C-(2-
cyanoethyl)thymidine (4). The NMR data (see Tables
1 and 2) for 4 supported the assigned ‘down’ stereo-
chemistry of the 30-substituent, as might be expected
from favored approach of the carbon addend on the face
opposite the large substituents at C-1 and C-4 of the
pentofuranose ring. The reaction is evidently stereo-
selective, but the possibility that the C-30 epimer might
have been present in the reaction mixture cannot be
ruled out.
Repetition of the reaction, but using thionocarbamate
3 as the radical precursor instead of 2, again gave a mix-
ture of products. However, in this instance, the yield of
crystalline 30-C-(2-cyanoethyl) derivative 4 was signifi-
cantly enhanced, to 52%, and it was concluded that thio-
nocarbamate precursor 3 constituted a more effective
radical precursor than thionoester 2.
A further evaluation of thionocarbamate 3 for C-
alkylation was conducted under similar conditions, with
methyl acrylate as the trapping agent. TLC of the reac-
tion mixture after removal of the 50-substituent by use of
tetrabutylammonium fluoride and chromatographic iso-
lation of the principal component gave 41% of a product
whose NMR data the indicated it to be 30-C-(2-meth-
oxycarbonyl)ethyl derivative (5) of thymidine.
Thiono groups serving as radical precursors for
subsequent C–C bond formation were introduced at
O-30 of the known11,15 50-O-tert-butyldimethylsilylthymi-
dine (1). The crystalline 30-(methylthio)thiocarbonyl
[CH3SC(@S)–] derivative (2) was obtained in 84% yield
by the successive treatment of 1 with sodium hydride, car-
bon disulfide, and methyl iodide, and the corresponding
30-(imidazol-1-yl)thiocarbonyl [(C3H3N2)C(@S)–] deriv-
ative (3) was isolated crystalline in 82% yield by treating
1 with 1,10-thiocarbonyldiimidazole.
A benzene solution of thionoester 2 and an excess of
acrylonitrile (a good radical trapping-agent), was heated
under argon, and tri-n-butyltin hydride and azobis(iso-
butanonitrile) (AIBN, as a radical-generating agent),
in benzene solution, was added at a controlled rate via
a syringe pump. Flash chromatography of the reaction
mixture revealed three main components. First eluted
was a product that appeared to be the carbonyl ana-
logue of thiono precursor 2, and this was followed by
the product of simple 30-deoxygenation of 2, identified
after the removal of the silyl group, as the known
20,30-dideoxy nucleoside. The third fraction, obtained
T
T
TBDMSOCH
2 O
TBDMSOCH
2 O
NaH, CS2, MeI
Thymidine
MeSCO
HO
S
1
Bu3SnH
AIBN
Im2CS
2
CH2=CHCN
T
T
T
T
TBDMSOCH
TBDMSOCH
TBDMSOCH
TBDMSOCH
2 O
2 O
2 O
2 O
AIBN
+
+
Bu3SnH
CH2=CHCN
MeSCO
ImCO
CH2CN
S
O
4
3
O
Me
HN
1. AIBN
T =
Bu3SnH
CH2=CHCO2Me
2. Bu4NF
O
N
T
HOCH2
O
TBDMS = ButMe2Si
(CH2)2
N
N
Im =
CO2Me
5

D. Horton et al. / Carbohydrate Research 342 (2007) 259–267
Table 1. 1H NMR chemical-shift dataa (d) for compounds 1–11
261
Compound
Chemical shifts, d
H-10
H-20
H-200
H-30
H-40
H-50
H-500
H-6
NH
1
2b
3
6.37 dd
6.43 dd
6.46 dd
6.09 dd
6.06 dd
6.39 dd
6.47 dd
6.52 dd
6.499 dd
6.57 dd
6.12 t
2.36 m
2.57 m
2.72 dd
2.25 m
2.11 m
2.20 m
2.29 m
2.13 m
4.46 m
5.95 d
5.93 d
2.43 m
4.04 dd
4.28 m
4.38 m
3.76 m
4.01 m
3.99 dd
4.27 dd
4.37 d
3.86 m
4.02 dd
4.06 dd
3.98 m
3.75 m
3.97 dd
4.09 dd
4.08 dd
7.50 d
8.73 s
3.91 dd
3.97 dd
3.76 m
4c
5
6
2.39 ddd
2.66 ddd
2.75 ddd
2.77 dd
2.77 dd
2.10 m
2.20 ddd
2.36 ddd
2.43 d
2.48 ddd
2.43 ddd
4.56 dd
6.15 dd
5.95 dd
6.09 dd
5.95 dd
5.05 m
3.85 dd
3.97 dd
4.08 dd
7.48 d
7.57 d
7.58 dd
7.59 d
7.61 d
7.52 d
8.41 br
8.34 s
8.64 s
9.17 br
9.68 br
8.83 s
7
8
9
4.36 d
4.39 d
4.06 dd
3.65 m
3.95 m
10
11
3.80–3.72 m
2.42 dt
a In CDCl3 with Me4Si as the internal standard.
b 2.57 (s, 3H, SMe).
c 2.35 (s, 3H, SMe).
of a known12,17 procedure, and 6 was further converted
into 30-thiono derivatives to serve as radical precursors
for replacement of the O-30 substituent by the allyl
group. The crystalline 30-O-(methylthio)thiocarbonate
(7) was obtained from 6 by successive treatment with
sodium hydride, carbon disulfide, and methyl iodide.
Three additional 30-thiono compounds were also pre-
pared for comparative evaluation in an effort to maxi-
mize yields of 30-C-alkylated products. Treatment of 6
with phenyl chlorothionoformate–pyridine afforded
crystalline 30-O-(phenoxy)thiocarbonate 8, previously
reported12 as a glass, in 96% yield. Similarly prepared
from 6 were the 30-O-(imidazol-1-yl)thiocarbamate 9,
obtained crystalline in 62% yield, and the 30-O-(penta-
fluorophenoxy)thiocarbonate 10, also crystalline, in
87% yield.
Each of the four compounds 7–10 was evaluated com-
paratively in reaction with allyltributyltin, with the
AIBN initiator generating tributyltin radicals that ab-
stract the thiono group at C-30 of the 50-protected nucleo-
side derivative to generate a radical species at C-30,
which in turn interacts with allyltributyltin to effect allyl-
ation at C-30.
The general procedure for the allylation reaction was
adapted from the techniques described by Keck et al.,18
with careful exclusion of oxygen. Each of the thiono pre-
cursors 7–10 reacted with allyltributyltin in boiling benz-
ene to produce a mixture of products from which the
desired 30-C-allyl product 11 could be isolated as a
low-melting crystalline solid, along with the product of
simple deoxygenation at C-30. Compound 11 was earlier
described10 as a foam, prepared from an o-tolyl ana-
logue of 8.
The results are shown in Table 3, where it may be seen
that the 1-[5-O-(tert-butyldiphenylsilyl)-2-deoxy-3-O-[(penta-
fluorophenoxy)thiocarbonyl]-b-D-erythro-pentofurano-
syl]thymine radical precursor 10 turned out to be the
best precursor among those examined, both in terms
of overall yield (89%) and reaction time (2 h). The yield
Table 2. 1H NMR spin-coupling constants (Hz) for compounds 1–11
Compound
Coupling constants (Hz)
0
0
0
00
0
00
0
0
0
0
0
00
0
00
J1 ;2
J1 ;2
J2 ;02
J3 ;4
J4 ;5
J4 ;5
J5 ;5
1
2
5.8
5.3
5.3
4.6
3.0
5.9
5.2
5.1
5.2
5.1
5.7
8.0
9.3
9.3
6.8
7.2
8.0
9.4
9.4
9.3
9.4
5.7
2.5
5.9
5.9
5.0
1.8
1.8
2.1
2.1
2.0
1.8
11.4
11.4
3
14.2
4
5
6a
7b
8c
9d
10e
11
8.5
14.0
14.1
14.4
14.1
14.5
2.9
2.0
1.1
3.6
12.2
11.5
6.0
3.9
5.9
3.5
11.5
a
b
c
0
00
0
J2;03 8:5; J2 ;3 2:5 Hz.
0
0
00
0
J2 ;3 5:2; J2 ;3 6:1 Hz.
00
0
J2 ;3 6:0 Hz.
d
e
00
0
J2 ;3 5:2 Hz.
00
0
J2 ;3 2:4 Hz.
Attempts to apply the foregoing free-radical proce-
dures using either maleimide, maleic anhydride, or phen-
ylacetylene as the radical trapper proved unsuccessful.
With a large excess of maleimide and an extended reac-
tion period, the starting compound was largely recov-
ered.16 The use of maleic anhydride again returned
mostly thionocarbamate 3, along with the product of
simple deoxygenation, the 50-protected 20,30-dideoxy-
thymidine (ddT) derivative. With phenylacetylene, the
only product detected was material formed by radical-
induced polymerization of phenylacetylene.16
The replacement of O-30 in the thymidine structure by
a C-allyl group via free-radical procedures has been re-
ported by Chu et al.,8 and Fiandor and Tam,9 using
the 30-O-phenoxythiocarbonyl derivative of compound
1, and by De Mesmaeker et al.10 using the tert-butyl-
diphenylsilyl protecting group and the 30-O-(o-tolyl-
oxy)thiocarbonyl group. Our studies employed the
tert-butyldiphenylsilyl protecting group, preparing from
thymidine the silyl ether 6 in 82% yield by a modification

262
D. Horton et al. / Carbohydrate Research 342 (2007) 259–267
T
T
TBDPSOCH2
TBDPSOCH2
O
O
ButPh2SiCl
NaH, CS2, MeI
Thymidine
HO
MeSCO
S
PhOCCl
S
7
6
C6F5OCSCl
Im2CS
T
T
T
T
TBDPSOCH2
TBDPSOCH2
TBDPSOCH2
TBDPSOCH2
O
O
O
O
CH2=CHCH2Bu3Sn
C6F5OCO
S
ImCO
CH2
PhOCO
S
S
CH=CH2
8
10
9
11
TBDPS = ButPh2Si
the parent 30-hydroxy compound 6; there was no evi-
dence for the formation of C-30 alkylated products.
In conclusion, this report demonstrates satisfactory
procedures for using acrylonitrile and methyl acrylate
as radical acceptors for the attachment of a carbon
chain at C-30 of a 50-protected thymidine via a 30-thiono
derivative, and an optimized procedure for C-30 allyl-
ation by the use of a 30-pentafluorophenoxythiono-
carbonate precursor. However maleimide, maleic
anhydride, and phenylacetylene failed to react under
comparable conditions, and the attempted introduction
of a vinyl or ethynyl group using tributylvinyltin or tri-
butylethynyltin was likewise unsuccessful.
Table 3. Conversion of 30-hydroxy precursor 6 into 30-C-allyl product
11
Protect
30-OH
Yield
(%)
Allylation
Yield
(%)
Overall
yield (%)
Total
time (h)
6 ! 7
6 ! 8
6 ! 9
6 ! 10
67
96
62
87
7 ! 11
8 ! 11
9 ! 11
10 ! 11
22
32
77
90
15
31
48
78
96
70
7
2
of C-allyl product 11 ranged from 22% for the (methyl-
thio)thiocarbonate 7, 32% for the (phenoxy)thiocarbon-
ate 8, and 72% for the (imidazol-1-yl)thiocarbamate 9,
and all required longer reaction times.
Further functional modification of the allyl group in
compound 11 to generate the formylmethyl derivative,
and its reduction to the corresponding C-(2-hydroxy-
methyl) derivative has been recorded elsewhere.19
Although there are conflicting reports as to whether
the 50-deprotected analogue of this derivative possesses
antiviral activity,20,21 it was disclosed20 to have anti-
tumor activity against four tumor cell lines (L1210,
P388, S-180, and CCRF-CEM).
Attempts to prepare 30-vinyl and 30-ethynyl analogues
of 30-allyl derivative 11 by similar free-radical proce-
dures were not successful.19 Although these analogues
are accessible7 by a long synthetic sequence, the current
approach offered promise for a simple and brief alterna-
tive. However, the procedure used for 30-C-allylation
of pentafluorophenoxythiocarbonyl derivative 10 in
reaction with allyltributyltin, repeated under a range
of conditions and variation of solvent with either
tributylvinyltin or tributylethynyltin led principally to
mixtures of the products of simple deoxygenation at
C-30 or removal of the thiono substituent to produce
1. Experimental
1.1. General methods
The solvents were purified and dried as recommended22
and the evaporation of solvents was performed under
diminished pressure at a bath temperature below
40 ꢁC, unless otherwise stated. Melting points were
determined with a Fisher–Johns melting point apparatus
without correction. Thin-layer chromatography (TLC)
was performed on Silica Gel 60 F254 (E. Merck) fol-
lowed by dipping the plates in 5% p-anisaldehyde soln,
with subsequent heating. Silica Gel 60 (E. Merck) was
used for column chromatography. Microanalyses were
performed by Atlantic Microlab Inc., Norcross, Ga.
1H NMR and 13C NMR spectra were recorded on the
following Bruker instruments, as noted in the tables:
AM-250 (250 MHz 1H, 62.5 MHz 13C); AM-300
(300 MHz 1H, 75 MHz 13C); WM-300 (300 MHz 1H,

D. Horton et al. / Carbohydrate Research 342 (2007) 259–267
263
75 MHz 13C); and AM-500 (500 MHz 1H, 125 MHz
13C). All spectra were recorded at ꢀ25 ꢁC with the spec-
trometers operating in the Fourier-transform mode. The
1H and 13C spectral assignments were made by compar-
ison with known compounds or with similar structures
whose spectral assignments have been published.
Decoupling techniques or COSY were also applied to
spectral assignments and to determine the coupling con-
stants as necessary. Mass spectra were recorded with
Kratos MS-30 or VG70-250s instruments at an ionizing
potential of 70 eV.
tracts were evaporated and the crude product was puri-
fied by flash-column chromatography (50:1 CHCl3–
MeOH). The fractions of major product (Rf 0.78, 10:1
CHCl3–MeOH) were collected and the solvent evapo-
rated to give a yellow syrup that was triturated with
petroleum ether. Filtration of the pale-yellow solid and
drying in vacuo at room temperature gave 2.94 g
(6.58 mmol, yield 84%) of compound 2 as white crystals;
1
mp 167–169 ꢁC; [a]D ꢁ38.0 (c 1.0, CHCl3); H NMR
(500 MHz, CDCl3): (see Tables 1 and 2); 13C NMR
(125.7 MHz, CDCl3): d 215.34 (s, C@S), 163.88 (s, C-
4), 150.51 (s, C-2), 134.88 (d, C-6), 111.22 (s, C-5),
85.02 (d, C-40), 84.76 (d, C-10), 83.90 (d, C-30), 63.64
(t, C-50), 38.08 (t, C-20), 25.85 [q, C(CH3)3], 19.20 (q,
SCH3), 18.24 (s, C(CH3)3), 12.41 (q, a-CH3); ꢁ5.44
(q, Si–CH3), ꢁ5.51 (q, Si–CH3); EIMS: m/z 447
[1.34, M+H]+, 339 [7.91, M+HꢁCH3SC(@S)OH]+,
321 [8.68, M+Hꢁthymine]+. Anal. Calcd for
C18H30N2O5S2Si (446.66): C, 48.40; H, 6.77; N,
6.27; S, 14.36. Found: C, 48.41; H, 6.82; N, 6.27; S,
14.46.
1.2. Preparation11,15 of 1-[5-O-(tert-butyldimethylsilyl)-2-
deoxy-b-D-erythro-pentofuranosyl]thymine (1)
tert-Butylchlorodimethylsilane (890 mg, 5.90 mmol) was
added to a soln of thymidine (1.38 g, 5.70 mmol) and
imidazole (780 mg, 11.5 mmol) in N,N-dimethylform-
amide (10 mL) at room temperature. After 2 h, the mix-
ture was poured into 50 mL of ice–water in a thin stream
with vigorous stirring, and then extracted with CH2Cl2
(50 mL · 2). The extracts were combined, dried
(MgSO4), and the solvent removed. The residue was trit-
urated with 15 mL of petroleum ether (bp 35–60 ꢁC), and
filtration gave 1.78 g of 1 as a white solid. The filtrate
was evaporated and the residue adsorbed onto 30 g of
silica gel and eluted with 10:1 toluene–1,4-dioxane.
Fractions containing the product of Rf 0.67 (1:1 1,4-
dioxane–benzene) were combined and after removal of
the solvent, afforded a further 80 mg of 1 as a white
solid. The combined yield was 1.86 g (5.22 mmol, 92%).
Recrystallization from CHCl3–C6H6 gave analytically
pure 1 as white crystals: mp 197–198.5 ꢁC; lit.11 mp
198–199 ꢁC; [a]D +1.65 (c 0.80, CHCl3); 1H NMR
(250 MHz, CDCl3): (see Tables 1 and 2); 13C NMR
(125.7 MHz, CDCl3): d 163.68 (C-4), 150.36 (C-2),
135.44 (C-6), 110.96 (C-5), 87.18 (C-40), 85.00 (C-10),
72.63 (C-30), 63.61 (C-50), 41.16 (C-20), 25.95
[C(CH3)3], 18.38 [C(CH3)3], 12.54 (a-CH3), ꢁ5.34 (Si–
CH3), ꢁ5.44 (Si–CH3).
1.4. 1-[5-O-(tert-Butyldimethylsilyl)-2-deoxy-3-O-[(imid-
azolyl)thiocarbonyl]-b-D-erythro-pentofuranosyl]thymine
(3)
A soln of compound 1 (5.70 g, 16.0 mmol) in anhyd
THF (15 mL) and 1,10-thiocarbonyldiimidazole (4.64 g,
90% purity, 23.4 mmol) under argon was boiled under
reflux for 2 h. The mixture was then evaporated and
the residue subjected to flash chromatography on silica
gel (50:1 CHCl3–MeOH) to give a light-yellow syrup
(Rf 0.61, 10:1 CHCl3–MeOH), which was triturated with
petroleum ether to yield 6.10 g (13.1 mmol, 82% yield)
of 3 as a white solid. Recrystallization from CHCl3
and petroleum ether gave 3 as white crystals; mp 105–
108 ꢁC; [a]D ꢁ44.0 (c 1.6, CHCl3); 1H NMR
(500 MHz, CDCl3): (see Tables 1 and 2); 13C NMR
(125.7 MHz, CDCl3): d 183.11 (s, C@S), 163.79 (s,
C-4), 150.56 (C-2), 137.16 (s, C-2 of imidazole), 134.53
(d, C-6), 131.10 (d, C-5 of imidazole), 117.81 (d, C-4
of imidazole), 111.46 (s, C-5), 84.83 (d, C-40), 84.80 (d,
C-10), 84.25 (d, C-30), 63.77 (t, C-50), 38.06 (t, C-20),
25.85 [q, C(CH3)3], 18.26 [s, C(CH3)3], 12.47 (q, a-
CH3), ꢁ5.40 (q, Si–CH3), ꢁ5.49 (q, Si–CH3).
1.3. 1-[5-O-(tert-Butyldimethylsilyl)-2-deoxy-3-O-[(meth-
ylthio)thiocarbonyl]-b-D-erythro-pentofuranosyl]thymine
(2)
A mixture of 1 (2.80 g, 7.86 mmol), NaH (80% disper-
sion in mineral oil, 5.60 g), and imidazole (10 mg) in
50 mL of dry tetrahydrofuran was stirred for 30 min at
room temperature until no more hydrogen was evolved.
Carbon disulfide (5.6 mL, 93 mmol) was added and stir-
ring continued for 1 h at room temperature, followed by
the addition of MeI (6.2 mL, 0.10 mol). The resultant
mixture was stirred for 30 min at room temperature
and then cooled in an ice–water bath. Water was added
to the mixture until no more hydrogen was evolved and
the product was extracted with 50 mL of Et2O. The ex-
1.5. 1-[5-O-(tert-Butyldimethylsilyl)-3-C-(2-cyanoethyl)-
2,3-dideoxy-b-D-erythro-pentofuranosyl]thymine (4)
1.5.1. From dithiocarbonate 2. A mixture of 500 mg
(1.12 mmol) of compound 2 and acrylonitrile (910 lL,
18.0 mmol) in anhyd benzene (20 mL) was heated
to reflux under argon. A soln of 0.45 mL (1.67 mmol)
of Bu3SnH and 40 mg of azobis(isobutanonitrile)
(AIBN) in 0.7 mL of anhydrous benzene was added
via a syringe pump at a rate of 0.2 mL/h. The resulting

264
D. Horton et al. / Carbohydrate Research 342 (2007) 259–267
mixture was boiled under reflux for an additional 2 h.
TLC analysis revealed that 2 was still present. Addi-
tional acrylonitrile (0.20 mL, 3.96 mmol) was added
and a soln (0.50 mL) of benzene containing 350 lL
(1.30 mmol) of Bu3SnH and 40 mg of AIBN was added
over 2 h, and boiling was continued for an additional
2 h. However, the amount of 2 present did not decrease
appreciably with this further addition. The mixture was
cooled to room temperature and evaporated to a syrup,
which was subjected to chromatography on silica gel
(100:1 CHCl3–MeOH). The first fraction crystallized
and was recrystallized from acetone–petroleum ether
to afford 1-[5-O-(tert-butyldimethylsilyl)-3-O-(methyl-
thio)carbonyl-2-deoxy-b-D-erythro-pentofuranosyl]thy-
mine as white crystals, yield 220 mg (45%), mp 160–
268 [31.41, M+Hꢁthymine]+. Anal. Calcd for
C19H31N3O4Si (393.56): C, 57.99; H, 7.94; N, 10.68.
Found: C, 57.77; H, 8.01; N, 10.53.
1.5.2. Compound 4 from thionocarbamate 3. To a soln
of 330 mg (0.71 mmol) of thionocarbamate 4 and acrylo-
nitrile (0.72 mL, 14.3 mmol) in 15 mL of dry benzene
boiled under reflux under argon was added a soln of
250 mg (0.93 mmol) of Bu3SnH and 12 mg of AIBN in
5.0 mL of anhydrous benzene over 5 h via a syringe
pump at the rate of 1.0 mL/h. The resulting soln was
boiled under reflux for an additional 2 h. TLC analysis
(10:1 CH3Cl–MeOH) revealed two major (Rf 0.68 and
0.58) and two minor products (Rf 0.43, 0.35). The sol-
vent was evaporated to give an oil that was resolved
by flash chromatography over silica gel. The fraction
corresponding to Rf 0.58 contained 4. The solvent was
evaporated and the residue triturated with petroleum
ether to give 147 mg (0.37 mmol, 52%) of compound 4
as a white solid.
25
1
162 ꢁC, ½aꢂD ꢁ0.5 (c 0.2, CHCl3); H NMR (500 MHz,
CDCl3): d 8.06 (s, 1H, NH), 7.51 (d, 1H, H-6), 6.35
(dd, 1H, J1 ;2 5:3; J1 ;2 9:3 Hz, H-10), 5.37 (d, 1H,
0
0
0
00
J2 ;3 H-30), 4.17 (d, 1H, J3 ;4 1:5 Hz, H-40), 3.91 (d,
2H, J 2.0 Hz, H-50,500), 2.47 (dd, 1H, Jgem 13.9 Hz, H-
20), 2.35 (s, 3H, COSCH3), 2.12 (m, 1H, H-200), 1.92
(d, 3H, a-CH3), 0.94 [s, 9H, C(CH3)3], 0.14 (s, 6H, Si–
Me); 13C NMR (125.7 MHz, CDCl3): d 171.63 (s,
OCOSCH3), 163.60 (s, C-4), 150.27 (s, C-2), 134.86 (d,
C-6), 111.17 (s, C-5), 85.04 (d, C-40), 84.61 (d, C-10),
78.13 (d, C-30), 63.41 (t, C-50), 38.00 (t, C-20), 25.84 [q,
C(CH3)3], 18.25 [s, C(CH3)3], 13.34 (s, SMe), 12.40 (q,
a-CH3), ꢁ5.46 (q, SiMe), ꢁ5.58 (q, SiMe); EIMS: m/z
431 (0.5, M+ꢁ1), 93 (100, CH3SCO2H++1). Anal. Calcd
for C18H30N2O6SSi (430.60): C, 50.20; H, 7.02; N, 6.50;
S, 7.45. Found: C, 50.32; H, 7.03; N, 6.55; S, 7.36.
The second fraction gave a crude white solid (26 mg,
7%), which was identified as 1-[5-O-(tert-butyldimethyl-
silyl)-2,3-dideoxy-b-D-erythro-pentofuranosyl]thymine,
as deprotection of the 50-OH group by treatment of a
soln in tetrahydrofuran with 1 M Bu4NF in tetrahydro-
furan for 2 h at room temperature gave 2,3-dideoxy-b-D-
erythro-pentofuranosylthymine; recrystallized from
acetone–petroleum ether it had mp 146–149 ꢁC (lit.23
0
0
0
0
1.6. 1-[2,3-Dideoxy-3-C-(2-methoxycarbonylethyl)-b-D-
erythro-pentofuranosyl]thymine (5) from thionocarbamate
3
To a soln of thionocarbamate 3 (1.01 g, 2.17 mmol) and
methyl acrylate (2.03 g, 23.6 mmol) in 10 mL of anhyd
benzene heated to reflux under argon was added a soln
of 1.17 mL (4.35 mmol) of Bu3SnH and 40 mg of AIBN
in 7.0 mL of anhyd benzene over a period of 8 h via a
syringe pump at a rate of 1 mL/h. The resultant soln
was boiled for an additional 2 h. TLC monitoring
(10:1 CHCl3–MeOH) of the reaction showed a major
product at Rf 0.64. After solvent removal, the residue
was subjected to flash chromatography over silica gel
(25:1 CHCl3–MeOH) to give the Rf 0.64 component
(0.930 g) as an oil. Analysis of this fraction revealed
the oil to still be a mixture. Deprotection was effected
by adding 2.0 mL (2 mmol) Bu4NF in 30 mL of THF
and stirring at room temperature for 30 min. TLC anal-
ysis revealed four products (Rf 0.39, 0.35, 0.32, and 0.27;
15:1 CHCl3–MeOH). Removal of solvent followed by
flash chromatography on silica gel (15:1 CHCl3–MeOH)
gave the homogeneous Rf 0.32 component 5 as a syrup
(350 mg, 0.89 mmol, 41%); [a]D ꢁ13 (c 0.5, CHCl3);
1H NMR (300 MHz, CDCl3): (see Tables 1 and 2);
13C NMR (500 MHz, CDCl3): d 173.43 (s, CO), 163.19
(s, C-4), 150.25 (s, C-2), 136.28 (d, C-6), 110.36 (s, C-
5), 88.32 (d, C-4), 85.26 (C-10), 61.75 (t, C-50), 51.82
(q, OCH3), 38.92 (t, C-20), 36.33 (d, C-30), 32.29 (t,
CH2CH2CO2CH3), 26.97 (t, CH2CH2CO2CH3), 12.52
25
mp 145 ꢁC), ½aꢂD +27 (c 0.62, MeOH); 13C NMR
(125.7 MHz, CDCl3): d 163.62 (C-4), 149.87 (C-2),
138.48 (C-6), 108.68 (C-5), 84.70 (C-10), 80.69 (C-40),
61.79 (C-50), 31.57 (C-20), 24.08 (C-30), 11.57 (a-CH3);
EIMS: m/z 227 (24, M++1), 126 (100, thymine++1),
101 (84, M++1ꢁthymine).
The third fraction (140 mg of a crude syrup) was trit-
urated with 10 mL of petroleum ether and 2 mL of Et2O
to yield 110 mg (0.28 mmol, 25%) of 4 as a white solid:
1
mp 117–120 ꢁC; [a]D +22.0 (c 0.25, CHCl3); H NMR
(500 MHz, CDCl3): (see Tables 1 and 2); 13C NMR
(125.7 MHz, CDCl3): d 163.59 (C-4), 150.26 (C-2),
135.23 (C-6), 118.61 (CN), 110.62 (C-5), 85.47 (C-40),
84.75 (C-10), 63.10 (C-50), 38.34 (C-20), 37.16 (CH2CN),
28.12 (CH2CH2CN), 25.95 [C(CH3)3], 18.46 [C(CH3)3],
15.97 (C-30), 12.60 (a-CH3), ꢁ5.36 (Si–CH3); EIMS:
m/z 394 [2.31, M+H]+, 336 [1.55, M+HꢁCH(CH3)3]+,
1
(a-CH3). Copies of the H and 13C NMR spectra of
compound 5 are recorded in the Supplementary data.
Repetition of this procedure under the same condi-
tions, but with maleimide in the place of methyl acrylate
showed no reaction, even after a second addition of

D. Horton et al. / Carbohydrate Research 342 (2007) 259–267
265
1
Bu3SnH and AIBN and further heating for 5 h in boil-
ing benzene, and starting compound 3 was recovered
in 84% yield. Similar results were observed when maleic
anhydride and phenylacetylene were used in the place of
methyl acrylate.16
CHCl3); H NMR (CDCl3): (see Tables 1 and 2); 13C
NMR (CDCl3):
d
215.32 (C@S), 163.99 (C-2),
150.64 (C-4), 135.65, 135.25, 134.89, 132.74, 131.87,
130.28, 130.16, 128.10 (aromatic carbons), 128.07
(C-6), 111.68 (C-5), 84.90 (C-40), 84.54 (C-10), 83.52
(C-30), 77.66 (SCH3), 64.38 (C-50), 38.17 (C-20), 27.03
[C(CH3)3], 19.39 [C(CH3)3], 12.08 (5-CH3); EIMS: m/z
571 [1.34, M+H]+, 513 [14.24, M+HꢁHC(CH3)3]+,
405 [48.19, M+HꢁCH3SC(@S)OHꢁHOC(CH3)3]+,
307 [21.47, M+HꢁCH3SC(@S)OHꢁ(C6H6)]+, 279
[79.36, M+HꢁCH3SC(@S)OHꢁ2(C6H6)ꢁthymine]+,
199 [51.06, (C6H5)2SiOH]+. Anal. Calcd for
C30H34N4O5SSi (570.807): C, 58.92; H, 6.00; N, 4.91.
Found: C, 58.63; H, 5.95; N, 4.90.
1.7. 1-[5-O-(tert-Butyldiphenylsilyl)-2-deoxy-b-D-erythro-
pentofuranosyl]thymine (6)
In a minor modification of the literature procedure,12,17
tert-butylchlorodiphenylsilane (8.375 mL, 31.85 mmol)
was added dropwise over 20 min to a 0 ꢁC soln of thy-
midine (6.24 g, 25.76 mmol) and imidazole (3.725 g,
54.5 mmol) in DMF (37.5 mL) under argon. The mix-
ture was stirred for 2 h and the solvent removed at
90 ꢁC. The resultant yellow syrup was dissolved in
CHCl3 (50 mL) and washed with water (40 mL · 2).
The CHCl3 layer was dried (Na2SO4) and evaporated
to furnish a colorless syrup, which was then dissolved
in 5:2 ether–MeOH, followed by the addition of hexane
(100 mL) with stirring. The resulting soln was cooled to
0 ꢁC, and a white crystalline product formed after 1 h.
The crystals were filtered off and washed with hexane
to provide a first crop of compound 6. A second crop
of 6 was obtained by concentration of the filtrate, fol-
lowed by the same purification procedure. The com-
bined yield was 10.22 g (82.5%): mp 162–164 ꢁC; lit.12
mp 162 ꢁC; Rf 0.38 (ether); [a]D +55.4 (c 1.0, CHCl3);
1H NMR (CDCl3): (see Tables 1 and 2); 13C NMR
(CDCl3): d 163.96 (C-2), 150.53 (C-4), 135.56, 135.45,
135.30, 132.95, 132.36, 130.16, 130.06, 128.01, 127.97
(aromatic carbons), 128.00 (C-6), 111.30 (C-5), 87.19
(C-40), 84.77 (C-10), 72.77 (C-30), 64.20 (C-50),
41.02 (C-20), 27.00 [C(CH3)3], 19.37 [C(CH3)3], 12.11
(5-CH3).
1.9. 1-[5-O-(tert-Butyldiphenylsilyl)-2-deoxy-3-O-[(phen-
oxy)thiocarbonyl]-b-D-erythro-pentofuranosyl]thymine (8)
To a suspension of 1-[5-O-(tert-butyldiphenylsilyl)-2-
deoxy-b-D-erythro-pentofuranosyl]thymine (6, 0.867 g,
1.80 mmol) in anhyd benzene (18 mL) was added phenyl
chlorothionoformate (0.249 mL, 1.80 mmol) with stir-
ring under argon at room temperature, and then after
2 min pyridine (0.147 mL, 1.82 mmol) was added in
one portion. The temperature was raised to boiling point
under reflux, and the reaction was complete after 2 h (a
white precipitate of pyridinium chloride formed). The
mixture was evaporated and the residue redissolved in
CHCl3 (30 mL). The soln was washed with water
(30 mL) and the aqueous layer extracted with CHCl3
(5 mL). The combined CHCl3 layers were dried (NaSO4)
and evaporated. Flash-column chromatography of the
crude product over silica gel with gradient elution (from
3:7 ether–hexane to pure ether) afforded, after evapora-
tion, compound 8 as a white crystalline product
(1.066 g, 1.728 mmol, 96%), mp 82.5–84 ꢁC; Rf 0.64
(ether); [lit.12 glass, Rf 0.79 (93:7 CH2Cl2–MeOH)]; [a]D
1.8. 1-[5-O-(tert-Butyldiphenylsilyl)-2-deoxy-3-O-[(meth-
ylthio)thiocarbonyl]-b-D-erythro-pentofuranosyl]thymine
(7)
1
+36.5 (c 1.0, CHCl3); H NMR (CDCl3): (see Tables 1
and 2); 13C NMR (CDCl3): d 194.23 (C@S), 153.29 (C-
2), 150.31 (C-4), 135.61, 135.23, 134.80, 132.67, 131.80,
130.25, 130.15, 129.62, 128.05 (aromatic carbons),
121.77 (C-6), 111.59 (C-5), 84.75 (C-40), 84.51 (C-10),
83.98 (C-30), 64.69 (C-50), 38.10 (C-20), 26.98
[C(CH3)3], 19.36 [C(CH3)3], 12.04 (5-CH3); EIMS: m/z
618 [0.02, M+2 H]+, 617 [0.01, M+H]+, 405 [5.62,
M+HꢁC6F5OC(@S)OHꢁHOC(CH3)3]+, 279 [22.14,
M+HꢁC6H5OC(@S)OHꢁHOC(CH3)3ꢁthymine]+, 94
[100, C6H5OH]+, 60 [46, O@C@S]+; lit.12 617 [M]+,
560 [M–C4H9]+.
To a suspension of 1-[5-O-(tert-butyldiphenylsilyl)-2-
deoxy-b-D-erythro-pentofuranosyl]thymine (6, 0.961 g,
2.00 mmol) in anhyd THF (7 mL) was added NaH
(87.3 mg, 3.64 mmol) with stirring under argon at room
temperature. After 30 min, CS2 (0.42 mL, 6.98 mmol)
was added in one portion, followed by MeI (0.28 mL,
4.45 mmol) 30 min later. The reaction was then
quenched after 20 min by adding AcOH until no more
hydrogen gas was evolved. The mixture was then diluted
with 7 mL of ether and filtered. Evaporation of the fil-
trate gave a yellowish syrup, which upon dissolution in
ether yielded a white crystalline solid. Purification of this
product on a column of silica gel by gradient elution
(from 1:1 ether–hexane to pure ether) afforded com-
pound 7 as white crystals (0.768 g, 1.35 mmol, 67%),
mp 133.0–134.5 ꢁC, Rf 0.89 (ether); [a]D +32.6 (c 1.0,
1.10. 1-[5-O-(tert-Butyldiphenylsilyl)-2-deoxy-3-O-
[(imidazolyl)thiocarbonyl]-b-D-erythro-pentofuranosyl]-
thymine (9)
To a suspension of 1-[5-O-(tert-butyldiphenylsilyl)-2-
deoxy-b-D-erythro-pentofuranosyl]thymine (6, 0.723 g,

266
D. Horton et al. / Carbohydrate Research 342 (2007) 259–267
1.50 mmol) in anhyd THF (15 mL) was added 1,10-thio-
carbonyldiimidazole (0.653 g, 2.97 mmol) with stirring
under argon at room temperature, and an additional
0.218 g (0.99 mmol) was added after the temperature
of the mixture was lowered to 20 ꢁC. The reaction was
complete after 12 h. Silica gel (5 g) was added to the
mixture to remove the brownish polar impurity and it
was then filtered off through a bed of silica gel. Evapo-
ration of the filtrate afforded 9 as a crystalline product
(0.554 g, 0.936 mmol, 62%): mp 174–175 ꢁC; Rf 0.89
(10:1 CHCl3–MeOH); [a]D +27.3 (c 1.0, CHCl3); 1H
NMR (CDCl3): (see Tables 1 and 2); 13C NMR
(CDCl3): d 182.94 (C@S), 163.79 (C-2), 150.60 (C-4),
137.12, 135.55, 135.18, 135.13, 134.44, 132.43, 131.68,
131.06, 130.30, 130.17, 128.07, 128.03 (aromatic and
imidazolyl carbons), 117.82 (C-6), 111.80 (C-5), 84.56
(C-40), 84.47 (C-10), 83.69 (C-30), 64.32 (C-50), 38.01
(C-20), 26.96 [C(CH3)3], 19.30 [C(CH3)3], 12.08
(5-CH3); EIMS: m/z 591 [0.04, M+H]+, 405 [33.80,
M+HꢁC3H3N2C(@S)OHꢁHC(CH3)3]+, 307 [29.65,
129.61, 128.05, 126.80 (aromatic carbons), 121.78
(C-6), 111.63 (C-5), 84.74 (C-40), 84.51 (C-10), 83.99
(C-30), 64.69 (C-50), 38.10 (C-20), 26.70 [C(CH3)3],
19.36 [C(CH3)3], 12.07 (5-CH3); EIMS: m/z 706 (0.06,
M+), 649 [11.11, M+HꢁHOC(CH3)3]+, 405 [28.43,
M+HꢁC6F5OC(@S)OHꢁHOC(CH3)3]+, 279 [56.39,
M+HꢁC6F5OC(@S)OHꢁHOC(CH3)3ꢁthymine]+, 184
[100, C6F5OH]+, 60 [46, O@C@S]+. Anal. Calcd for
C33H31F5N2O6SSi (706.77): C, 56.08; H, 4.42; N,
3.97. Found: C, 56.19; H, 4.48; N, 3.89.
1.12. 3-C-Allyl-1-[5-O-(tert-butyldiphenylsilyl)-2,3-dide-
oxy-b-D-erythro-pentofuranosyl]thymine (11) from pen-
tafluorophenoxythionocarbonate 10
The general procedure of Keck et al.18 for radical
allylation was followed with modifications. A soln
of 1-[5-O-(tert-butyldiphenylsilyl)-2-deoxy-3-O-[(penta-
fluorophenoxy)thiocarbonyl]-b-D-erythro-pentofuranos-
yl]thymine
(10,
706.7 mg,
1.000 mmol)
and
M+HꢁC3H3N2C(@S)OHꢁ2(C6H6)]+,
279
[81.45,
allyltributyltin (1.55 mL, 5.00 mmol) in anhyd benzene
(30 mL) was degassed with stirring under vacuum at
room temperature, and then charged with argon.
The procedure was repeated three times before adding
AIBN (90 mg, 0.55 mmol) in one portion under an
argon atmosphere. The temperature was increased to
reflux and the reaction was complete after 2 h. Two
products were visible by TLC analysis. After removal
of the solvent at 65 ꢁC, the colorless syrup was
chromatographed on a silica gel column via gradient
elution (from 1:1 ether–hexane to pure ether) to
provide a minor fraction of 1-[5-O-(tert-butyldiphenyl-
silyl)-2,3-dideoxy-b-D-erythro-pentofuranosyl]thymine
and a major fraction of 1-[3-C-allyl-5-O-(tert-butyl-
diphenylsilyl)-2,3-dideoxy-b-D-erythro-pentofuranosyl]-
thymine, which after evaporation of the eluate, was
obtained as a white crystalline product (449 mg,
89.5%): mp 45.0–46.5 ꢁC; Rf 0.72 (ether); [a]D +40.8
(c 1.0, CHCl3); 1H NMR (CDCl3): (see Tables 1
and 2); 13C NMR (CDCl3): d 163.90 (C-2), 150.41
(C-4), 135.53, 135.40, 135.36, 135.31, 133.17, 132.76,
130.00, 129.92, 128.91, 127.86, 127.84, 127.54
(aromatic carbons, C-200 and C-300), 117.06 (C-6),
110.57 (C-5), 85.53 (C-40), 84.71 (C-10), 63.85 (C-50),
38.17 (C-20), 37.13 (C-30), 36.44 (C-100), 27.00
[C(CH3)3], 19.38 [C(CH3)3], 12.15 (5-CH3); EIMS: m/z
506 [0.15, M+2H]+, 447 [14.33, M+HꢁHC(CH3)3]+,
M+HꢁC3H3N2C(@S)OHꢁHC(CH3)3ꢁthymine]+, 81
[100.00, C5H5O]+. Anal. Calcd for C30H34N4O5SSi
(590.807): C, 61.00; H, 5.80; N, 9.48. Found: C, 60.83;
H, 5.82; N, 9.44.
1.11. 1-[5-O-(tert-Butyldiphenylsilyl)-2-deoxy-3-O-
[(pentafluorophenoxy)thiocarbonyl]-b-D-erythro-pento-
furanosyl]thymine (10)
To a suspension of 1-[5-O-(tert-butyldiphenylsilyl)-2-
deoxy-b-D-erythro-pentofuranosyl]thymine (6, 2.89 g,
6.0 mmol) in anhyd benzene (100 mL) at room temper-
ature was added pentafluorophenyl chlorothionofor-
mate (1.10 mL, 6.6 mmol) in one portion. Pyridine
(0.54 mL, 6.7 mmol), which had been dried over
NaOH, was added after stirring for 2 min. The temper-
ature was raised to 80 ꢁC under an argon atmosphere
(a white precipitate of pyridine hydrochloride formed).
The reaction was complete after refluxing for 2 h. The
mixture was evaporated at 70 ꢁC to furnish a porous
solid, which was washed with water (50 mL · 3) and
then dissolved in CHCl3 (80 mL) and washed with
water (50 mL) again. The separated water layer was
again extracted with CHCl3 (20 mL). The combined
CHCl3 layers were dried (Na2SO4) and evaporated to
afford a foamy solid. The solid was dissolved in ether
and subjected to flash-column chromatography over
silica gel with gradient elution (pure hexane to pure
ether). Evaporation of the eluate at 70 ꢁC for 2 h,
and then at room temperature for 20 h, afforded 10
as a white solid (3.72 g, 0.525 mmol, 87%): mp 80–
321
[100.00,
M+HꢁHC(CH3)3ꢁthymine]+,
199
[49.26, M+Hꢁ(C6H5)2SiOH]+, 105 [7.12, C6H5Si]+.
Anal. Calcd for C29H36N2O4Si (504.703): C, 69.01;
H, 7.19; N, 5.55. Found: C, 69.06; H, 7.21; N,
5.56.
25
1
82 ꢁC; Rf 0.79 (ether); ½aꢂD +32.0 (c 1.0, CHCl3); H
NMR (CDCl3): (see Tables 1 and 2); 13C NMR
(CDCl3): d 194.25 (C@S), 153.30 (C-2), 150.53 (C-4),
135.61, 135.23, 134.79, 132.67, 131.82, 130.24, 130.15,
Repetition of this procedure using thiono derivatives
7, 8, and 9 gave inferior yields of 11 accompanied by
the product of simple deoxygenation at C-30 and its 30-
hydroxy analogue; details are given elsewhere.19

D. Horton et al. / Carbohydrate Research 342 (2007) 259–267
267
1.13. Attempted synthesis of 30-vinyl and 30-ethynyl
analogues of 11
Wolfrom, M. L.; Garg, H. G.; Horton, D. J. Org. Chem.
1965, 30, 1556–1560.
3. Sarafianos, S. G.; Hughes, S. H.; Arnold, E. Int. J.
Biochem. Cell Biol. 2004, 36, 1706–1715.
Repetition of the procedure used for the conversion of
thionoester 10 into 30-C-allyl product 11, but using tri-
butylvinyltin instead of allyltributyltin and xylene as
the solvent led19 to a mixture of five components, none
of which was the desired 30-C-vinyl derivative. Similar
results were observed using acetonitrile as the solvent
and also with 8 as the precursor and benzene or xylene
as the solvent. Likewise, the use of tributylethynyltin
under conditions for the conversion of 10 into 11, with
xylene or acetonitrile as the solvent gave mainly the
product of deesterification at C-30, and the use of 8 as
the precursor gave a similar result.19
4. De Clercq, E. Nature Rev. Drug Discovery 2002, 1, 13–25.
5. For preliminary reports, see: Lee, H. C., Chen, K., No, Z.,
Horton, D., Abstr. Pap., 20th Int. Carbohydr. Symp.,
Hamburg, Germany, August 27–September 1, 2000,
Abstr. B-160; Abstr. Pap. Am. Chem. Soc. Meet. 2000,
220, CARB-76 (2000), Washington, DC, August 20–24,
2000.
6. Herdewijn, P.; Balzarini, L.; De Clercq, E.; Pauwels, R.;
Baba, M.; Broder, S.; Vanderhaeghe, H. J. Med. Chem.
1987, 30, 1270–1278.
7. Sahlberg, C. Tetrahedron Lett. 1992, 33, 679–682.
8. Chu, C. K.; Doboszewski, B.; Schmidt, W.; Ullas, G. V.;
Van Roey, P. J. Org. Chem. 1989, 54, 2767–2769.
9. Fiandor, J.; Tam, S. Y. Tetrahedron Lett. 1990, 31, 597–
600.
10. De Mesmaeker, A. D.; Lebreton, J.; Waldner, A.;
Fristsch, V.; Wolf, R. M.; Freier, S. M. Synlett 1993,
733–736.
Supplementary data
11. Ogilvie, K. K. Can. J. Chem. 1973, 51, 3799–3807.
12. Sanghvi, Y. S.; Bharadwaj, R.; Debart, F.; De Mesmae-
ker, A. D. Synthesis 1994, 1163–1166.
13. Keck, G.; Yates, J. B. J. Am. Chem. Soc. 1982, 104, 5829–
5831.
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.carres.
2006.11.015.
14. Giese, B.; Dupuis, J.; Leising, M.; Nix, M.; Lindner, H. J.
Carbohydr. Res. 1987, 171, 329–341.
15. Ogilvie, K. K.; Iwacha, D. J. Tetrahedron Lett. 1973, 317–
319.
References
16. No, Z. Ph.D. Dissertation, Ohio State University, 1990;
Dissertation Abstr. Int. 1991, 51, 5876.
1. Simons, C. Nucleoside Mimetics; Gordon & Breach
Scientific, 2001; Chapter 1.
17. Matulie-Adamie, J.; Watananbe, K. A. J. Chem. Soc.,
Chem. Commun. 1985, 1535.
18. Keck, G. E.; Enholm, E. J.; Yates, J. B.; Wiley, M. R.
Tetrahedron 1985, 41, 4079–4094.
19. Chen, K. Ph.D. Dissertation, Ohio State University, 1996;
Dissertation Abstr. Int. 1997, 57, 4393.
20. Lin, T. S.; Zhu, J. L.; Dutschman, G. E.; Cheng, Y. C.;
Prusoff, W. H. J. Med. Chem. 1993, 36, 353–362.
21. Svansson, L.; Kvarnstro¨m, I.; Classon, B.; Samuelsson, B.
J. Org. Chem. 1991, 56, 2993–2997.
2. Vejcik, S. M.; Horton, D. Carbohydr. Res., in press;
doi:10.1016/j.carres.2006.12.003; Norris, P.; Horton, D.;
Levine, B. R. Heterocycles 1996, 43, 2643–2655; Blieszner,
K. C.; Horton, D.; Markovs, R. A. Nucleic Acid Chem-
istry: New and/or Improved Synthetic Procedures, Methods,
and Techniques Part 3; Wiley, 1986; pp 135–139; Horton,
D.; Liu, C.-M. Carbohydr. Res. 1982, 109, 282–289;
Horton, D.; Liu, C.-M. Carbohydr. Res. 1982, 107, 55–
70; Horton, D.; Markovs, R. A. Carbohydr. Res. 1980, 80,
356–363; Horton, D.; Kokrady, S. S. Carbohydr. Res.
1980, 80, 364–374; Blieszner, K. C.; Horton, D.; Markovs,
R. A. Carbohydr. Res. 1980, 80, 24l–262; Horton, D.;
Markovs, R. A. Carbohydr. Res. 1980, 80, 262–275;
22. Gordon, A. J.; Ford, R. A. The Chemist’s Companion;
John Wiley & Sons, 1972; p 379.
23. Julien, P. H.; Moffatt, J. G. US Patent 3,282,941, 1964.