Synthesis and structural elucidation of 1-(3-C-ethynyl-4-thio-β-D-ribofuranosyl)cytosine (4′-thioECyd)

Article abstract of DOI:10.1039/b204993g

A practical synthesis of 1,4-anhydro-4-thio-D-ribitol (5) via 1,4-dibromo-1,4-dideoxy-2,3,5-tri-O-benzyl-L-lyxitol (12) is described. This method reduced our previous eleven step procedure starting from D-ribose by three steps. The Pummerer reaction of 1,

Full text of DOI:10.1039/b204993g

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis and structural elucidation of 1-(3-C-ethynyl-4-thio-ꢀ-D-
ribofuranosyl)cytosine (4Ј-thioECyd)
Noriaki Minakawa,a Daisuke Kaga,a Yuka Kato,a Kanji Endo,b Motohiro Tanaka,c
Takuma Sasakic and Akira Matsuda*a
1
a Graduate School of Pharmaceutical Sciences, Hokkaido University, Kita-12, Nishi-6, Kita-ku,
Sapporo 060-0812, Japan
b Hanno Research Center, Taiho Pharmaceutical Co Ltd., 1–27 Misugidai Hanno,
Saitama 357–8527, Japan
c Cancer Research Institute, Kanazawa University, Takara-machi 13–1, Kanazawa 920–0934,
Japan
Received (in Cambridge, UK) 22nd May 2002, Accepted 6th August 2002
First published as an Advance Article on the web 3rd September 2002
A practical synthesis of 1,4-anhydro-4-thio--ribitol (5) via 1,4-dibromo-1,4-dideoxy-2,3,5-tri-O-benzyl--lyxitol (12)
is described. This method reduced our previous eleven step procedure starting from -ribose by three steps. The
Pummerer reaction of 1,4-anhydro-2-O-(2,4-dimethoxybenzoyl)-3,5-O-(1,1,3,3-tetraisopropyldisiloxane-1,3-diyl)-4-
sulfinyl--ribitol (6) in the presence of N 4-benzoylcytosine afforded the 4Ј-thiocytidine derivative 7b on a large scale.
Starting with the resulting 7b, 1-(3-C-ethynyl-4-thio-β--ribofuranosyl)cytosine (4; 4Ј-thioECyd), of which the 4Ј-oxo
congener ECyd (3) is a new type of potent antineoplastic nucleoside developed in our group, was synthesized via
elaborate protection and deprotection procedures, and successive reaction with cerium trimethylsilylacetylide. X-Ray
crystal structures of 4Ј-thioECyd (4) and ECyd (3) are presented in this paper. Although striking differences in bond
lengths and angles were observed in C1Ј–S4Ј and C4Ј–S4Ј, and C4Ј–S4Ј–C1Ј, the overall structures of each
compound, including the sugar puckering mode and the syn/anti conformation around the glycosyl bond, were
similar.
phosphate, which is successively converted to the active metab-
olite, ECyd 5Ј-triphosphate. The resulting 5Ј-triphosphate
Introduction
The 4Ј-thionucleosides, in which the furanose ring oxygen is
replaced by a sulfur atom, have been of particular interest
strongly inhibits cell growth and induces apoptosis as an RNA
polymerase inhibitor.6 Thus, ECyd is expected to be a new type
because of their biological properties, such as antiviral and
of antitumor nucleoside, and is currently under investigation in
antitumor activities.1 One of the most straightforward methods
Phase I clinical trials.
in the design of new biologically interesting 4Ј-thionucleosides
is to target bioisosteric compounds of the biologically active
4Ј-oxonucleosides. Since 4Ј-thionucleosides show different
susceptibility to enzymes of nucleoside metabolism, this tactic
In view of the above background, we decided to synthesize
1-(3-C-ethynyl-4-thio-β--ribofuranosyl)cytosine (4; 4Ј-thio-
ECyd) and to evaluate its antitumor activity. In contrast to the
synthesis and biological evaluation of the 2Ј-deoxy-4Ј-thio-
cytidine derivatives, none of the 4Ј-thiocytidine derivatives have
as yet been synthesized. This consideration also prompted
sometimes gives 4Ј-thio congeners which are more biologically
active than the parent 4Ј-oxonucleosides.2 For example, 4Ј-thio-
2Ј-deoxy-2Ј-methylidenecytidine (2; 4Ј-thioDMDC)3 showed
us to commence this investigation. We have already reported
higher antineoplastic activity than 2Ј-deoxy-2Ј-methylidene-
the stereoselective synthesis of 4Ј-thioribonucleosides via the
cytidine (1; DMDC), which has been synthesized in our group
Pummerer reaction.7 In this method, the reaction is carried out
(Fig. 1).4
using protected sulfoxide 6, prepared from -ribose via 1,4-
anhydro-4-thio--ribitol (5), and a silylated nucleobase to give a
4Ј-thioribonucleoside, such as the 4Ј-thiocytidine derivative 7a
(Scheme 1). Since the resulting 7 seemed a promising starting
Fig. 1
As part of our ongoing program for synthesizing new bio-
logically active nucleosides, we have reported the synthesis and
potent antitumor activities of 1-(3-C-ethynyl-β--ribofurano-
syl)cytosine (3; ECyd).5 ECyd first undergoes phosphorylation
by uridine-cytidine kinase (UCK) to give ECyd 5Ј-mono-
Scheme 1
2182
J. Chem. Soc., Perkin Trans. 1, 2002, 2182–2189
This journal is © The Royal Society of Chemistry 2002
DOI: 10.1039/b204993g

View Article Online
material to synthesize the desired 4, a large scale synthesis of 7
was required. To achieve this, the development of an improved
synthesis of 5 from -ribose was essential. Thus, we first
investigated the practical synthesis of 5, and succeeded in
reducing our previous eleven step synthesis by three steps. In
addition, this method is suitable for large scale production.
Consequently, compound 7b was obtained in multi tens of
grams using the Pummerer reaction. In this paper, we describe
in detail the improved synthesis of 5 and conversion of 7b to the
desired 4. We discuss the X-ray structures of ECyd (3) and 4Ј-
thioECyd (4) and their cytotoxicity against tumor cells in vitro.
Results and discussion
Scheme 3
As summarized in Scheme 2, 1,4-anhydro-4-thio--ribitol (5)
was originally prepared on a multigram scale from -ribose in
eleven steps.7b However, inversion of the secondary hydroxy
group of 8 by the Mitsunobu reaction, followed by cyclization
to give 11 makes this a tedious method. Therefore, we
attempted to simplify the synthetic route to 5 from 8.
Nagasawa, et al. reported that α,ω-dibromoalkanes cyclize
effectively to give cyclic sulfides by treatment with sodium
sulfide.8 Thus, we thought that 11 could be prepared in fewer
steps if we could substitute not only the primary but also
the secondary hydroxy group by bromide with inversion of
stereochemistry to give 12.
Scheme 4 Reagents: (i) MsCl, pyridine; (ii) LiBr, MEK, reflux;
(iii) Na2Sؒ9H2O, DMF, 100 ЊC; (iv) BCl3, CH2Cl2, Ϫ78 ЊC.
mechanism was then examined. Lithium bromide and tetra-
butylammonium bromide were tested as bromination reagents
in various solvents such as acetonitrile, DMF, 1,3-dimethyl-
imidazolidin-2-one and methyl ethyl ketone (MEK). Among
the attempts, the best result was obtained when 14 was treated
with ten equiv. of well-dried lithium bromide in MEK under
reflux conditions to give 12 in 56% yield. The resulting 12
cyclized effectively to give 11, as in the case of 10. As a practical
large scale synthesis, 11 was prepared from 8 in 65% yield
in three steps without purification of the intermediates 14 and
12 (see Experimental Section). To conduct these successive
modifications, it should be noted that TLC analysis of the
bromination is important for the increased yield of 11. The
monobrominated derivative 15 is first observed in the early
period of the reaction and is gradually converted to the less
polar product 12. In the synthesis of 11 from 8 on large scale,
however, a small portion of the undesired 17 (less than 10%)
was obtained along with 11 as an inseparable mixture, despite
the fact that TLC confirmed the disappearance of the
dimesylate 14 and the monobrominated derivative 15. This
could be attributed to further substitution of 12 by the bromo-
nium ion to give 16. Since 16 was not separable from 12 by
TLC, it is recommended that one quenches the reaction
immediately after the disappearance of 15. Fortunately, the
undesired 17 could be removed later during the conversion to
give 6, the substrate for the Pummerer reaction. The conditions
for debenzylation of 11 were also improved. In our previous
Scheme 2
Based on these considerations, a direct substitution of the
hydroxy groups with bromide was first examined. When com-
pound 8 was treated with NBS in the presence of triphenyl-
phosphine,9 however, 2,3,5-tri-O-benzyl-1,4-anhydro--ribitol
(13)10 was obtained as the major product (81% yield) along
with 12% yield of the desired dibromo derivative 12. It is
thought that this result is due to the low reactivity of the
secondary hydroxy group relative to the primary hydroxy
group. Consequently, a nucleophilic attack of the oxygen of
the secondary hydroxy group on the activated 1-position took
place prior to the desired substitution (Scheme 3). Since no
other conditions to give 12 preferentially in one-pot were
found, a stepwise method was next examined. As shown in
Scheme 4, compound 8 was converted to the dimesylate 14 in
95% yield by treatment with methanesulfonyl chloride in
pyridine. Interestingly, in contrast, treatment of 8 with toluene-
p-sulfonyl chloride or 4-nitrobenzenesulfonyl chloride under
the same conditions afforded 13 as the sole product, but not
the desired disulfonate. Nucleophilic bromination via an SN2
J. Chem. Soc., Perkin Trans. 1, 2002, 2182–2189
2183

View Article Online
Table 1 Geometric parameters: bond lengths, angles, and torsion
angles that represent important structural features of 4 and 3a
method, the reaction was carried out at less than Ϫ90 ЊC, which
was difficult to control on large scale. A reaction temperature of
at least Ϫ78 ЊC is required. As a result, 5 was obtained in 68%
yield when a solution of 11 in CH2Cl2 was added dropwise to a
precooled solution of BCl3 in CH2Cl2. Consequently, we
developed the practical large scale synthesis of 5 from 8, and
succeeded in paring three steps from the original protocol (44%
yield in four steps).
4Ј-thioECyd (4)
ECyd (3)
Bond lengths/Å
C1Ј–C2Ј
C2Ј–C3Ј
C3Ј–C4Ј
C1Ј–S4Јb
C4Ј–S4Јb
C1Ј–N1
1.527 (3)
1.532 (3)
1.546 (3)
1.818 (2)
1.840 (3)
1.473 (3)
1.523 (3)
1.551 (3)
1.547 (4)
1.408 (3)
1.449 (3)
1.458 (3)
After 5 was converted into 6,7b the Pummerer reaction was
performed on 6 in the presence of N 4-benzoylcytosine on a
large scale to give a sufficient amount of 7b for further use. An
effective method for introducing an ethynyl group on the C-3Ј
position of nucleosides stereoselectively has been reported
Jung et al. involving the reaction of 2Ј-O-TBDMS-3Ј-
ketonucleosides with cerium (trimethylsilyl)acetylide,11 the
utility of which has also been confirmed by us.12 This method
should be applicable to 4Ј-thio congeners, and thus, conversion
of 7b into the 2Ј,5Ј-di-O-TBDMS derivative 23 was examined.
Since attempts for selective deprotection of the 2,4-dimethoxy-
benzoyl (DMBz) group on the 2Ј-hydroxy group were unsuc-
cessful, both acyl protective groups were removed at once.
When 7b was treated with methylamine in MeOH solution, 18
was obtained in 69% yield. When methanolic ammonia was
used instead for this deacylation, a longer reaction time was
required and resulted in a decreased yield of 18. Selective
benzoylation of the N-4 amino group was achieved by treat-
ment of 18 with benzoic anhydride in DMF to give 19. The silyl
protective groups on the 3Ј and 5Ј-hydroxy groups were easily
removed to give N 4-benzoyl-4Ј-thiocytidine (20). Hakimelahi et
al. reported selective silylation of ribonucleosides to give 2Ј,5Ј-
di-O-silylated derivatives promoted by silver nitrate.13 When 20
was treated with TBDMSCl in the presence of silver nitrate,
however, the 5Ј-monoprotected compound was obtained, but
none of the disilylated derivative (data not shown). All attempts
at selective silylation of 20 did not work well. The desired 23
was effectively synthesized by elaborate protection and depro-
tection procedures. Thus, silylation of the 2Ј-hydroxy group of
19 was carried out by treatment with TBDMSOTf in the pres-
ence of 2,6-lutidine to give 21. This reaction did not proceed
under the usual conditions such as TBDMSCl and imidazole
in DMF. The selective desilylation of 1,1,3,3-tetraisopropyl-
disiloxane-1,3-diyl (TIPDS) group was achieved to give 22,
when 21 was treated with tetrabutylammonium fluoride in the
presence of an equimolar amount of acetic acid. Migration of
the remaining TBDMS group to the adjacent 3Ј-hydroxy group
was not observed under the conditions. As a practical syn-
thesis, 22 was prepared from 19 without purification of 21 in
73% yield in 2 steps. Silylation of the 5Ј-hydroxy group by
TBDMSOTf gave the desired 2Ј,5Ј-di-O-silylated derivative 23
(Scheme 5).
Bond angles (deg)
C1Ј–C2Ј–C3Ј
C2Ј–C3Ј–C4Ј
C3Ј–C4Ј–S4Јb
C4Ј–S14–C1Јb
S4Ј–C1Ј–C2Јb
S4Ј–C1Ј–N1b
108.6 (2)
107.2 (2)
106.3 (2)
94.9 (1)
107.4 (2)
112.2 (2)
101.7 (2)
101.4 (2)
106.1 (2)
110.7 (2)
106.3 (2)
109.1 (2)
Torsion angles (deg)
C4Ј–S4Ј–C1Ј–C2Јb (ν0)
S4Ј–C1Ј–C2Ј–C3Јb (ν1)
C1Ј–C2Ј–C3Ј–C4Ј (ν2)
C2Ј–C3Ј–C4Ј–S4Јb (ν3)
C3Ј–C4Ј–S4Ј–C1Јb (ν4)
S4Ј–C1Ј–N1–C2b (χ)
O5Ј–C5Ј–C4Ј–C3Ј (γ)
Ϫ10.1 (2)
31.7 (2)
Ϫ42.8 (3)
33.7 (2)
Ϫ13.7 (2)
Ϫ134.6 (2)
58.1 (3)
Ϫ19.5 (3)
34.5 (2)
Ϫ35.2 (2)
25.0 (2)
Ϫ4.0 (3)
Ϫ139.2 (2)
58.6 (3)
Pseudorotation parameters (deg)
Phase angle (P)
Puckering amplitude (νm)
182.2
42.8
167.5
36.1
a Sds (standard errors) are given in parentheses. b S represents O4Ј in the
case of compound 3.
Compound 23 was then oxidized with DMSO–Ac2O to give
the 3Ј-keto derivative 24. To direct the preferential nucleophilic
addition to the ketone from the β face, assistance of a free
hydroxy group at the 5Ј-position was needed.11 Treatment of 24
with 90% aqueous TFA gave the hydroxy ketone 25. Immedi-
ately after the usual water work-up, 25 was successively treated
with cerium (trimethylsilyl)acetylide to give 26 in 64% yield in 2
steps stereoselectively, as in the case of the 4Ј-oxo derivatives.
Deprotection of 26 by ammonium fluoride, followed by methyl-
amine in MeOH, gave the desired product 4 quantitatively
(Scheme 6).
The structure of 4 was confirmed by X-ray analysis (Fig. 2).
As can be seen in Fig. 2a, the ethynyl group is introduced on the
β-face of the thio sugar. To date, several X-ray structures of
4Ј-thionucleosides have been presented,14 and it was revealed
that a marked conformational change in the carbohydrate
ring compared with the corresponding 4Ј-oxonucleoside was
observed despite the resemblance of their overall structures,
namely the sugar puckering mode and the syn/anti conform-
ation around the glycosyl bond. To elucidate the structural
Fig. 2 The crystal structures of (a) 4Ј-thioECyd (4) and (b) ECyd (3).
differences, an X-ray structure of 3 is also presented in Fig. 2b.
In Table 1, important conformational characteristics for the
structures 3 and 4 are summarized. Among these data, striking
differences in the bond lengths and angles were observed in
2184
J. Chem. Soc., Perkin Trans. 1, 2002, 2182–2189

View Article Online
Scheme 5 Reagents: (i) N 4-benzoylcytosine, TMSOTf, Et3N, CH2Cl2–toluene; (ii) MeNH2 in MeOH (40%); (iii) Bz2O, DMF, 50 ЊC; (iv) TBAF,
AcOH, THF; (v) TBDMSOTf, 2,6-lutidine, CH2Cl2.
P = 167.5Њ and νm = 36.1Њ, respectively. This type of conform-
ation was also maintained in both structures in solution. The
conformation of each sugar rings was analyzed on the basis of
the coupling constant J1Ј,2Ј in DMSO-d6.15 The J value of 4
was found to be 8.6 Hz, while that of 3 was 6.6 Hz. These data
show that both compounds prefer a South-type puckered
conformation in solid state and in solution.
In spite of the overall structural resemblance of 4 to 3, how-
ever, 4 did not show any significant cytotoxicity against L1210
and KB cells in vitro at the concentration of 100 µg mLϪ1, while
3 was a strong inhibitor of tumor cell proliferation with IC50
values of 16 and 28 nM, respectively, against the same cell
lines.5 One explanation for this result would be the difference of
susceptibility to nucleoside and/or nucleotide kinase(s). As
described in the introduction, 3 is phosphorylated by UCK to
give its 5Ј-monophosphate, which is successively converted to
the active metabolite, ECyd 5Ј-triphosphate.6 In these metabolic
activations, the first phosphorylation is thought to be the most
important step. Thus, we tested the first phosphorylation of 4
by partially purified UCK from mouse Sarcoma-180 cells, and
compared the relative susceptibility to the enzyme with 3 and
the natural substrate, cytidine. Consequently, 3 was phos-
phorylated 26% relative to cytidine,12 while phosphorylation of
4 was not detected under the same conditions (data not shown).
To our knowledge, only one other 4Ј-thioribonucleoside, that is,
4Ј-thioadenosine has been tested for susceptibility to bovine
liver adenosine kinase, and it was revealed that this nucleoside
was a poor substrate for the kinase.16 As was the case for
4Ј-thioadenosine, the 4Ј-thiocytidine derivative, 4 was also a
poor substrate for UCK. To date, potent biological activities
of several 2Ј-deoxy-4Ј-thionucleoside derivatives have been
reported.2,3 These results strongly suggest that some 2Ј-deoxy-
4Ј-thionucleosides would be good substrates for kinases such as
thymidine kinase and deoxycytidine kinase. In contrast to
the deoxynucleoside kinases, ribonucleoside kinases may dis-
criminate structural changes in the sugar conformations, and
do not metabolize 4Ј-thioribonucleosides to the corresponding
5Ј-phosphates.
Scheme 6 Reagents: (i) Ac2O, DMSO; (ii) aq. TFA; (iii) lithium
acetylide, CeCl3, THF, Ϫ78 ЊC; (iv) NH4F, MeOH, reflux, then MeNH2
in MeOH (40%).
C1Ј–S4Ј and C4Ј–S4Ј, and C4Ј–S4Ј–C1Ј. Thus, the bond
lengths C1Ј–S4Ј and C4Ј– S4Ј of 4 were 1.818 (2) and 1.840 (3)
Å, respectively, while the corresponding bond lengths of 3 were
much shorter (i.e., 1.408 (3) and 1.449 (3) Å, respectively). The
other bond lengths including the glycosidic bond (C1Ј–N1)
were quite similar to those of 3. In contrast to the longer bond
length of 4, the bond angle C4Ј–S4Ј–C1Ј in the thio sugar is
94.9Њ, which is 15.8Њ less than that of 3. The other bond angles
in the two sugar moieties do not differ markedly. In spite of the
partial structural differences between 3 and 4, their overall
structures are quite similar. Thus, the cytosine bases are both in
the anti conformation with the glycosidic torsion χ(S4Ј–C1Ј–
N1–C2) = Ϫ134.6Њ and χ(O4Ј–C1Ј–N1–C2) = Ϫ139.2Њ. The
4Ј-thiosugar of 4 is found in a South-type puckered conform-
ation with pseudorotation phase angle P = 182.2Њ and
maximum puckering amplitude νm = 42.8Њ. The furanose ring of
3 also exhibits a South-type conformation with the values of
In conclusion, we have developed an improved synthesis of
the 1,4-anhydro-4-thio--ribitol (5), in which we succeeded in
reducing our previous eleven step synthesis by three steps. This
method facilitates a large scale synthesis of the 4Ј-thiocytidine
derivative 7b, which is thought to be a good substrate for
J. Chem. Soc., Perkin Trans. 1, 2002, 2182–2189
2185

View Article Online
Physical data for 1,4-dibromo-1,4-dideoxy-2,3,5-tri-O-benzyl-L-
lyxitol (12)
further derivatized 4Ј-thiocytidine analogs. Starting with the
resulting 7b, 4Ј-thioECyd (4) was synthesized via elaborate pro-
tection and deprotection procedures. In addition, X-ray crystal
structural analyses of 4 and 3 have been done. The present
results indicate that the overall structures of 4 and 3 are similar.
Contrary to our expectations, 4Ј-thioECyd (4) showed no
significant cytotoxicity against L1210 and KB cells in vitro.
Further investigation is needed to elucidate the effect of
the sulfur atom at the 4Ј-position of the nucleosides and their
susceptibility to nucleoside and/or nucleotide kinases. The
improved method presented in this work should be the driving
force for further studies.
Found: C, 56.88; H, 5.15. C26H28Br2O3 requires C, 56.95; H,
5.15%; 1H NMR (400 MHz, CDCl3) δ 7.40–7.27 (m, 15 H, Ph),
4.80–4.43 (m, 7 H), 3.99–3.67 (m, 6 H); 13C NMR (100 MHz,
CDCl3) δ 137.57, 137.23, 137.09, 128.30, 128.09, 128.01, 127.90,
127.86, 127.71, 127.60, 127.55, 77.89, 76.62, 74.94, 73.06, 72.09,
70.70, 53.39, 33.40 FAB-LRMS m/z 547, 549, and 551 (MHϩ,
MHϩ ϩ 2, and MHϩ ϩ 4, 2.6, 3.8 and 1.6%).
Physical data for 1,4-anhydro-2,3,5-tri-O-benzyl-4-thio-L-lyxitol
(17)
1H NMR (270 MHz, CDCl3) δ 7.34–7.25 (m, 15 H, Ph), 4.87
and 4.69 (each d, each 1 H, CH2Ph, J = 11.7 Hz), 4.55 and 4.48
(each s, each 2 H, CH2Ph × 2), 4.19 (m, 1 H, H-3), 4.05 (ddd,
1 H, H-2, J = 2.6, 6.2, 9.1 Hz), 3.91 (m, 1 H, H-4), 3.08 (dd, 1 H,
Experimental section
General methods
Physical data were measured as follows. Melting points are
uncorrected. 1H and 13C NMR spectra were recorded at 270 or
400 MHz and 100 MHz instruments in CDCl3 or DMSO-d6 as
the solvent with tetramethylsilane as an internal standard.
Chemical shifts are reported in parts per million (δ), and signals
are expressed as s (singlet), d (doublet), t (triplet), m (multiplet),
or br (broad). All exchangeable protons were detected by
addition of D2O. Mass spectra were measured on JEOL JMS-
D300 spectrometer. X-Ray measurements were made on
AFC-7R with graphite-monochromated Mo-Kα radiation or
AFC-5R with graphite-monochromated CuKa radiation,
Rigaku. TLC was done on Merck Kieselgel F254 precoated
plates. Silica gel used for column chromatography was Merck
silica gel 5715.
H-1a, J = 9.4, 9.7 Hz), 2.91 (dd, 1 H, H-1b, J = 6.2, 9.7 Hz); 13
C
NMR (100 MHz, CDCl3) δ 138.47, 137.97, 137.90, 128.29,
128.22, 128.11, 127.63, 127.56, 127.49, 127.38, 127.25, 83.40,
78.71, 73.49, 73.23, 72.05, 70.13, 45.64, 30.31; FAB-LRMS m/z
421 (MHϩ, 12.5%); FAB-HRMS 421.1850 (MHϩ, C26H29O3S
requires m/z 421.1837).
1,4-Anhydro-4-thio-D-ribitol (5)7b
A solution of 1 M BCl3 in CH2Cl2 (1.49 L, 1.49 mol) was
cooled to Ϫ78 ЊC, and a solution of 11 (104.5 g, 0.24 mol)
in CH2Cl2 (1 L) was added to the precooled solution over
3.5 h. After the solution was stirred at the same temperature for
2 h, the reaction was quenched by addition of a mixture of
MeOH–CH2Cl2 (2 : 1, 900 mL) at Ϫ78 ЊC. The solvent was
removed in vacuo, and the residue was coevaporated with
MeOH. The residue was purified by a silica gel column, eluted
with 4–16% MeOH in CHCl3, to give 5 (25.5 g, 68% as a yellow
oil).
1,4-Anhydro-2,3,5-tri-O-benzyl-4-thio-D-ribitol (11)7b
To a solution of 8 (248.9 g, 0.59 mol) in dry pyridine (1 L) was
added methanesulfonyl chloride (114 mL, 1.47 mol) at 0 ЊC.
After the mixture was stirred for 1 h at the same temperature,
the reaction was quenched by addition of ice. The reaction mix-
ture was partitioned between AcOEt and H2O. The separated
organic layer was washed with saturated aqueous NaHCO3,
followed by brine. The organic layer was dried (Na2SO4) and
concentrated in vacuo, and the residue was coevaporated with
toluene to give 14. The crude 14 was dissolved in MEK (1.2 L),
and well-dried lithium bromide (512 g, 5.89 mol) was added to
the solution. The mixture was heated under reflux for 7 h. After
being cooled to room temperature, the mixture was diluted with
AcOEt and washed with H2O, followed by brine. The organic
layer was dried (Na2SO4) and concentrated in vacuo to give
crude 12. The resulting 12 was dissolved in dry DMF (1.2 L),
and sodium sulfide nanohydrate (142 g, 0.59 mol) was added to
the solution. The mixture was heated at 100 ЊC for 1 h. After
being cooled to room temperature, the mixture was diluted with
AcOEt and washed with H2O, followed by brine. The organic
layer was dried (Na2SO4) and concentrated in vacuo. The resi-
due was purified by a silica gel column, eluted with hexane–
AcOEt (50 : 1–10 : 1), to give 11 including less than 10% of 17
(162 g, 65% in three steps).
N 4-Benzoyl-1-[2-O-(2,4-dimethoxybenzoyl)-3,5-O-(1,1,3,3-
tetraisopropyldisiloxane-1,3-diyl)-4-thio-ꢀ-D-ribofuranosyl]-
cytosine (7b)
To a suspension of N 4-benzoylcytosine (14.4 g, 67.0 mmol) in
dry toluene (350 mL) was added triethylamine (9.3 mL, 67.0
mmol) and TMSOTf (51.8 mL, 268.1 mmol), and the mixture
was stirred at room temperature until giving a two-phase clear
solution. Dry CH2Cl2 (200 mL) was added to the above solu-
tion, which gave an one-phase clear solution, and the whole was
added to a solution of 6 (25.6 g, 44.7 mmol) in dry CH2Cl2 (200
mL) dropwise over 30 min via a cannula. Additional triethyl-
amine (28.0 mL, 201.1 mmol) in dry toluene (100 mL) was
added dropwise to the reaction mixture at 0 ЊC to initiate the
Pummerer reaction. After being stirred for 15 min at room tem-
perature, the reaction was quenched by addition of ice, and the
reaction mixture was partitioned between AcOEt and H2O. The
separated organic layer was washed with saturated aqueous
NaHCO3, followed by brine. The organic layer was dried
(Na2SO4) and concentrated in vacuo. The residue was purified
by a silica gel column, eluted with hexane–AcOEt (3 : 1–1 : 3),
to give 7b (22.9 g, 67% as a white solid): Found: C, 57.76; H,
6.68; N 5.36. C37H51N3O9SSi2 requires C, 57.71; H, 6.68; N,
5.46%; 1H NMR (400 MHz, CDCl3) δ 8.67 (br s, 1 H, NH), 8.61
(d, 1 H, H-6, J = 7.9 Hz), 7.92–7.85 (m, 3 H, o-DMBz, o-Bz),
7.65–7.50 (m, 4 H, H-5, m-Bz, p-Bz), 6.52–6.48 (m, 2 H,
m-DMBz), 6.16 (s, 1 H, H-1Ј), 5.72 (d, 1 H, H-2Ј, J = 4.0 Hz),
4.49 (dd, 1 H, H-3Ј, J = 4.0, 9.2 Hz), 4.18 (dd, 1 H, H-5Јa,
J = 3.0, 12.9 Hz), 4.09 (d, 1 H, H-5Јb, J = 12.9 Hz), 3.86 (s, 6 H,
MeO × 2), 3.78 (dd, 1 H, H-4Ј, J = 9.2, 3.0 Hz), 1.14–0.87 (m, 28
H, TIPDS); 13C NMR (100 MHz, CDCl3) δ 166.34, 164.10,
163.09, 161.84, 161.24, 154.75, 145.99, 133.78, 133.09, 132.85,
128.91, 127.44, 112.01, 104.45, 98.91, 96.56, 77.45, 71.42, 64.18,
Physical data for 1,4-bis(O-methanesulfonyl)-2,3,5-tri-O-benzyl-
D-ribitol (14)
Found: C, 58.13; H, 5.94. C28H34O9S2 requires C, 58.11; H,
5.92%; 1H NMR (270 MHz, CDCl3) δ 7.35–7.24 (m, 15 H, Ph),
5.13 (ddd, 1 H, H-4, J = 7.5, 3.0, 3.2 Hz), 4.76–4.44 (m, 7 H,
H-1a, CH2Ph × 3), 4.32 (dd, 1 H, H-1b, J = 11.3, 4.0 Hz), 3.91
(dd, 1 H, H-3, J = 3.0, 7.5 Hz), 3.77–3.71 (m, 2 H, H-2, H-5a),
3.59 (dd, 1 H, H-5b, J = 3.2, 11.1 Hz), 3.00 and 2.93 (each s,
each 3 H, Me); 13C NMR (100 MHz, CDCl3) δ 137.10, 136.70,
136.58, 128.39, 128.32, 128.23, 128.13, 128.00, 127.77, 127.59,
81.42, 77.25, 75.89, 73.89, 72.30, 68.57, 67.49, 38.62, 37.66;
FAB-LRMS m/z 579 (MHϩ, 12.4%).
2186
J. Chem. Soc., Perkin Trans. 1, 2002, 2182–2189

View Article Online
N 4-Benzoyl-1-[2-O-(tert-butyldimethylsilyl)-4-thio-ꢀ-D-ribo-
furanosyl]cytosine (22)
58.12, 56.01, 55.53, 50.90, 17.62, 17.56, 17.52, 17.18, 17.02,
17.00, 13.51, 13.31, 13.27, 12.73; FAB-LRMS m/z 770 (MHϩ,
8.3%).
To a solution of 19 (1.8 g, 3.0 mmol) in dry CH2Cl2 (15 mL) was
added 2,6-lutidine (3.5 mL, 30.0 mmol) and TBDMSOTf
(3.4 mL, 15.0 mmol) at 0 ЊC. The mixture was stirred for 15 h at
room temperature. The reaction was quenched by addition of
1 M aqueous HCl, and the mixture was stirred for 10 min. The
reaction mixture was partitioned between AcOEt and H2O. The
separated organic layer was washed with saturated aqueous
NaHCO3, followed by brine. The organic layer was dried
(Na2SO4) and concentrated in vacuo to give crude 21. Com-
pound 21 was dissolved in THF (30 mL), and AcOH (0.34 mL,
6.0 mmol) and TBAF (1 M in THF, 6.0 mL, 6.0 mmol)
were added to the mixture at 0 ЊC. After being stirred for 2 h
at the same temperature, the reaction mixture was par-
titioned between AcOEt and H2O. The separated organic layer
was washed with H2O, followed by brine. The organic layer
was dried (Na2SO4) and concentrated in vacuo. The residue was
purified by a silica gel column, eluted with hexane–AcOEt
(2 : 1–1 : 1), to give 22 (1.04 g, 73% as a white solid): 1H NMR
(400 MHz, CDCl3) δ 9.00 (br s, 1 H, NH), 8.66 (d, 1 H, H-6,
J = 7.5 Hz), 7.87 (d, 2 H, o-Bz, J = 7.5 Hz), 7.61–7.46 (m, 4 H,
H-5, m-Bz, p-Bz), 5.87 (d, 1 H, H-1Ј, J = 3.6 Hz), 4.52 (dd, 1 H,
H-2Ј, J = 3.6, 3.4 Hz), 4.17 (m, 1 H, H-3Ј), 4.08 (dd, 1 H, H-5Јa,
J = 3.0, 11.7 Hz), 3.93 (dd, 1 H, H-5Јb, J = 3.2, 11.7 Hz), 3.75
(br s, 1 H, 5Ј-OH), 3.58 (ddd, 1 H, H-4Ј, J = 6.2, 3.0, 3.2 Hz),
2.82 (d, 1 H, 3Ј-OH, J = 5.8 Hz), 0.90 (s, 9 H, t-Bu), 0.18, 0.11
(each s, each 3 H, Me × 2); 13C NMR (100 MHz, CDCl3)
δ 166.50, 161.96, 155.05, 147.50, 133.10, 132.75, 128.88, 127.47,
96.90, 79.13, 74.78, 69.25, 62.13, 52.63, 25.84, 18.10, Ϫ4.45,
Ϫ4.93; FAB-LRMS m/z 478 (MHϩ, 25.5%); FAB-HRMS
478.1830 (MHϩ, C22H32N3O5SSi requires m/z 478.1832).
1-[3,5-O-(1,1,3,3-Tetraisopropyldisiloxane-1,3-diyl)-4-thio-ꢀ-D-
ribofuranosyl]cytosine (18)
Compound 7b (1.4 g, 1.82 mmol) was dissolved in methylamine
in MeOH solution (40%, 90 mL), and the mixture was kept for
3 h at room temperature. The solvent was removed in vacuo, and
the residue was coevaporated with MeOH. The residue was
purified by a silica gel column, eluted with 4% MeOH in
CHCl3, to give 18 (0.63 g, 69% as a white foam): Found: C,
50.19; H, 7.73; N, 8.16. C21H39N3O5SSi2 requires C, 50.27;
H, 7.83; N, 8.37%; 1H NMR (400 MHz, DMSO-d6) δ 8.14
(d, 1 H, H-6, J = 7.3 Hz), 7.17 (d, 2 H, NH2), 5.86 (d, 1 H,
2Ј-OH, J = 4.1 Hz), 5.68 (d, 1 H, H-5, J = 7.3 Hz), 5.58 (s,
1 H, H-1Ј), 4.00–3.89 (m, 4 H, H-2Ј, H-3Ј, H-5Ј), 3.52 (m, 1 H,
H-4Ј), 1.10–0.88 (m, 28 H, TIPDS); 13C NMR (100 MHz,
CDCl3) δ 165.52, 156.37, 142.28, 94.44, 78.63, 71.75, 65.85,
58.51, 49.75, 17.64, 17.58, 17.53, 17.51, 17.22, 17.20, 17.17,
17.06, 13.46, 13.29, 13.24, 12.48; FAB-LRMS m/z 502 (MHϩ,
68.8%).
N 4-Benzoyl-1-[3,5-O-(1,1,3,3-tetraisopropyldisiloxane-1,3-diyl)-
4-thio-ꢀ-D-ribofuranosyl]cytosine (19)
To a solution of 18 (6.9 g, 13.8 mmol) in DMF (140 mL) was
added Bz2O (4.7 g, 20.6 mmol), and the whole was stirred at 50
ЊC for 9 h. The reaction was quenched by addition of saturated
aqueous NaHCO3 at 0 ЊC, and the reaction mixture was par-
titioned between AcOEt and H2O. The separated organic layer
was washed with saturated aqueous NaHCO3, followed by
brine. The organic layer was dried (Na2SO4) and concentrated
in vacuo. The residue was purified by a silica gel column, eluted
with hexane–AcOEt (3 : 1–2 : 1), to give 19 (7.8 g, 93% as a
white foam): Found: C, 55.37; H, 7.15; N, 6.71. C28H43N3O6SSi2
N 4-Benzoyl-1-[2,5-bis-O-(tert-butyldimethylsilyl)-4-thio-ꢀ-D-
ribofuranosyl]cytosine (23)
1
To a solution of 22 (48.2 mg, 0.101 mmol) in dry CH2Cl2 (1 mL)
was added 2,6-lutidine (70.5 µL, 0.60 mmol) and TBDMSOTf
(69.5 µL, 0.30 mmol) at 0 ЊC. The mixture was stirred for 30 min
at the same temperature. The reaction was quenched by
addition of 1 M aqueous HCl, and the mixture was stirred for
10 min. The reaction mixture was partitioned between AcOEt
and H2O. The separated organic layer was washed with satur-
ated aqueous NaHCO3, followed by brine. The organic layer
was dried (Na2SO4) and concentrated in vacuo. The residue was
purified by a silica gel column, eluted with hexane–AcOEt
(4 : 1–3 : 1), to give 23 (52.9 mg, 89% as a white foam): Found:
C, 56.71; H, 7.58; N, 7.10. C28H45N3O5SSi2 requires C, 56.82;
requires C, 55.50; H, 7.15; N, 6.94%; H NMR (270 MHz,
CDCl3) δ 8.75–8.72 (m, 2 H, NH, H-6, J = 7.3 Hz), 7.92–7.89
(m, 2 H, o-Bz), 7.65–7.49 (m, 4 H, H-5, m-Bz, p-Bz), 5.97 (s, 1
H, H-1Ј), 4.28 (dd, 1 H, H-3Ј, J = 3.3, 9.2 Hz), 4.23 (d, 1 H, H-
2Ј, J = 3.3 Hz), 4.15 (dd, 1 H, H-5Јa, J = 3.0, 12.9 Hz), 4.06 (d, 1
H, H-5Јb, J = 12.9 Hz), 3.72 (dd, 1 H, H-4Ј, J = 9.2, 3.0 Hz),
2.90 (br s, 1 H, 2Ј-OH), 1.14–0.87 (m, 28 H, TIPDS); 13C NMR
(100 MHz, CDCl3) δ 165.93, 161.95, 155.25, 146.33, 133.13,
132.76, 129.76, 128.94, 128.14, 127.41, 96.28, 78.82, 72.03,
66.13, 58.42, 50.12, 17.61, 17.57, 17.50, 17.22, 17.17, 17.15,
17.05, 13.52, 13.31, 13.28, 12.57; FAB-LRMS m/z 606 (MHϩ,
19.8%).
1
H, 7.66; N, 7.10%; H NMR (400 MHz, CDCl3) δ 8.82 (d,
N 4-Benzoyl-1-(4-thio-ꢀ-D-ribo-pentofuranosyl)cytosine (20)
1 H, H-6, J = 7.0 Hz), 8.67 (br s, 1 H, NH), 7.90 (d, 2 H, o-Bz,
J = 7.3 Hz), 7.70–7.49 (m, 4 H, H-5, m-Bz, p-Bz), 6.05 (d, 1 H,
H-1Ј, J = 2.6 Hz), 4.29 (dd, 1 H, H-2Ј, J = 2.6, 3.2 Hz), 4.11 (m,
1 H, H-3Ј), 4.06 (dd, 1 H, H-5Јa, J = 2.9, 11.1 Hz), 3.92 (dd, 1 H,
H-5Јb, J = 2.6, 11.1 Hz), 3.51 (ddd, 1 H, H-4Ј, J = 6.4, 2.9,
2.6 Hz), 2.29 (d, 1 H, 3Ј-OH, J = 8.2 Hz), 0.98, 0.94 (each s, each
9 H, t-Bu × 2), 0.27, 0.18, 0.17, 0.15 (each s, each 3 H, Me × 4);
13C NMR (100 MHz, CDCl3) δ 166.41, 161.69, 155.01, 146.69,
132.99, 128.82, 128.60, 127.43, 96.54, 80.19, 73.53, 66.91, 62.04,
52.43, 26.12, 25.84, 18.78, 18.10, Ϫ4.37, Ϫ5.01, Ϫ5.12, Ϫ5.15;
FAB-LRMS m/z 592 (MHϩ, 7.3%).
To a solution of 19 (306 mg, 0.50 mmol) in THF (10 mL)
was added AcOH (57 µL, 1.0 mmol) and TBAF (1 M in THF,
1.0 mL, 1.0 mmol) at 0 ЊC. After being stirred for 10 min at
the same temperature, the solvent was removed in vacuo. The
residue was suspended in EtOH, and collected by filtration.
The solid was washed by EtOH, and dried to give 20 (164 mg,
89% as a white solid): Found: C, 52.82; H, 4.96; N, 11.28.
1
C16H17N3O5S requires C, 52.88; H, 4.72; N, 11.56%; H NMR
(270 MHz, DMSO-d6) δ 11.24 (br s, 1 H, NH), 8.57 (d, 1 H,
H-6, J = 7.9 Hz), 8.00 (d, 2 H, o-Ar, J = 7.3 Hz), 7.64–7.47
(m, 3 H, m-Ar, p-Ar), 7.38 (d, 1 H, H-5, J = 7.9 Hz), 5.96
(d, 1 H, H-1Ј, J = 5.9 Hz), 5.53 (d, 1 H, 2Ј-OH, J = 5.9 Hz), 5.24
(d, 1 H, 3Ј-OH, J = 4.6 Hz), 5.18 (t, 1 H, 5Ј-OH, J = 5.3 Hz),
4.20 (m, 1 H, H-2Ј), 4.03 (m, 1 H, H-3Ј), 3.72 (ddd, 1 H, H-5Јa,
J = 5.9, 11.2, 5.3 Hz), 3.60 (ddd, 1 H, H-5Јb, J = 4.6, 11.2, 5.3
Hz), 3.28 (m, 1 H, H-4Ј); 13C NMR (100 MHz, DMSO-d6)
δ 167.19, 162.59, 155.19, 146.74, 133.01, 132.72, 128.41, 96.80,
77.20, 72.98, 64.43, 62.91, 53.08; FAB-LRMS m/z 364 (MHϩ,
16.7%).
N 4-Benzoyl-1-[2,5-bis-O-(tert-butyldimethylsilyl)-4-thio-ꢀ-D-
ribofuran-3-ulosyl]cytosine (24)
A mixture of 23 (48.4 mg, 0.08 mmol) and Ac2O (0.4 mL)
in DMSO (0.8 mL) was stirred at room temperature for 2.5 h.
The reaction was quenched by addition of saturated aqueous
NaHCO3, and the whole was stirred for 10 min. The mixture
was partitioned between AcOEt and H2O. The separated
organic layer was washed with saturated aqueous NaHCO3,
J. Chem. Soc., Perkin Trans. 1, 2002, 2182–2189
2187

View Article Online
followed by brine. The organic layer was dried (Na2SO4) and
concentrated in vacuo. The residue was purified by a silica gel
column, eluted with hexane–AcOEt (4 : 1–3 : 1), to give 24 (41.4
mg, 86% as a white foam): Found: C, 56.71; H, 7.29; N, 7.04.
C28H43N3O5SSi2 requires C, 57.01; H, 7.35; N, 7.12%; 1H NMR
(400 MHz, CDCl3) δ 8.65 (br s, 1 H, NH), 8.32 (d, 1 H, H-6,
J = 7.3 Hz), 7.90 (m, 2 H, o-Bz), 7.64–7.51 (m, 4 H, H-5,
m-Bz, p-Bz), 6.22 (d, 1 H, H-1Ј, J = 7.3 Hz), 4.36 (d, 1 H, H-2Ј,
J = 7.3 Hz), 4.18 (dd, 1 H, H-5Јa, J = 4.7, 11.4 Hz), 3.89–3.86
(m, 2 H, H-4Ј, H-5Јb), 0.95, 0.85 (each s, each 9 H, t-Bu × 2),
0.15, 0.12, 0.08, Ϫ0.01 (each s, each 3 H, Me × 4); 13C NMR
(100 MHz, CDCl3) δ 203.98, 166.12, 161.53, 155.02, 145.15,
133.20, 132.69, 128.92, 127.45, 97.67, 80.40, 63.75, 59.57, 52.20,
26.06, 25.53, 18.64, 18.14, Ϫ4.46, Ϫ5.11, Ϫ5.15, Ϫ5.18;
FAB-LRMS m/z 590 (MHϩ, 6.7%).
δ 7.97 (d, 1 H, H-6, J = 7.3 Hz), 7.15 (br d, 2 H, NH2), 5.97 (d,
1 H, H-1Ј, J = 8.6 Hz), 5.93 (s, 1 H, 3Ј-OH), 5.76 (d, 1 H, H-5,
J = 7.3 Hz), 5.58 (d, 1 H, 2Ј-OH, J = 7.9 Hz), 5.10 (dd, 1 H, 5Ј-
OH, J = 4.6, 5.3 Hz), 4.22 (dd, 1 H, H-2Ј, J = 8.6, 7.9 Hz), 3.79–
3.76 (m, 2 H, H-5Јa, H-5Јb), 3.54 (s, 1 H, 3Ј-C᎐CH), 3.20 (dd,
᎐
1 H, H-4Ј, J = 4.0, 5.3 Hz); 13C NMR (100 MHz, CDCl3)
δ 165.00, 155.76, 144.37, 94.84, 83.50, 80.39, 77.59, 74.87,
63.43, 61.49, 55.59; FAB-LRMS m/z 284 (MHϩ, 22.9%).
X-Ray crystallography†
Crystal data for 4: C11H13N3O4S, M = 283.30, orthorhombic,
a = 10.887 (3), b = 10.892 (3), c = 10.385 (3) Å, V = 1231.4 (5) Å,3
T = 296 K, space group P212121 (no. 19), Z = 4, µ(MoKα) = 2.78
cmϪ1, 1655 reflections measured, 1637 unique (Rint = 0.000)
which were used in all calculations. The final R was 0.037
(I > 2.0σ (I)).
N 4-Benzoyl-1-[2-O-(tert-butyldimethylsilyl)-3-C-(trimethyl-
silyl)ethynyl-4-thio-ꢀ-D-ribofuranosyl]cytosine (26)
Crystal data for 3: C11H13N3O5, M = 267.24, orthorhombic,
a = 10.657 (4), b = 10.894 (3), c = 10.362 (3) Å, V = 1202.9 (5) Å,3
T = 296 K, space group P212121 (no. 19), Z = 4, µ(Cu-Kα) =
A solution of 24 (628 mg, 1.06 mmol) in an aqueous TFA solu-
tion (90%, 5 mL) was stirred for 20 min at 0 ЊC. The reaction
mixture was added to saturated aqueous NaHCO3 at 0 ЊC, and
the whole was stirred for 10 min. The reaction mixture was
partitioned between AcOEt and H2O. The separated organic
layer was washed with saturated aqueous NaHCO3, followed by
brine. The organic layer was dried (Na2SO4) and concentrated
in vacuo. The residue was coevaporated with toluene to give
crude 25 as a white solid. The unstable hydroxy ketone 25 was
used directly without further purification.
10.12 cmϪ1, 2273 reflections measured, 2249 unique (Rint
=
0.026) which were used in all calculations. The final R was 0.048
(I > 4.0σ (I)).
Acknowledgements
This investigation was supported in part by a Grant-in-Aid for
Scientific Research on Priority Areas from the Ministry of
Education, Science, Sports, and Culture of Japan. We would
like to thank Ms H. Matsumoto and Ms A. Maeda (Center for
Instrumental Analysis, Hokkaido University) for elemental
analysis. We also would like to thank Ms S. Oka (Center for
Instrumental Analysis, Hokkaido University) for measurement
of Mass spectra.
To a solution of (trimethylsilyl)acetylene (0.9 mL, 6.39
mmol) in dry THF (5 mL) was added n-BuLi (15% w/v in
hexane, 4.0 mL, 6.39 mmol) at Ϫ20 ЊC. The mixture was stirred
for 1 h at the same temperature. The resulting lithium acetylide
solution was added to the anhydrous CeCl3 suspension (1.57 g,
6.39 mmol) in THF (5 mL) via a cannula at Ϫ78 ЊC. The mix-
ture was stirred for 1 h at the same temperature, and a solution
of crude 25 in dry THF (5 mL) was added via a cannula at
Ϫ78 ЊC. After being stirred for 30 min at Ϫ78 ЊC, the reaction
was quenched by addition of AcOH (0.9 mL), and the temper-
ature was raised to room temperature. The mixture was par-
titioned between AcOEt and H2O. The separated organic layer
was washed with H2O, followed by brine. The organic layer was
dried (Na2SO4) and concentrated in vacuo. The residue was
purified by a silica gel column, eluted with hexane–AcOEt
(4 : 1–2 : 1), to give 26 (390 mg, 64% as a white foam): 1H NMR
(270 MHz, CDCl3) δ 8.91 (br s, 1 H, NH), 8.22 (d, 1 H, H-6,
J = 6.9 Hz), 7.91 (d, 2 H, o-Bz, J = 7.6 Hz), 7.61–7.48 (m, 4 H,
H-5, m-Bz, p-Bz), 5.92 (d, 1 H, H-1Ј, J = 5.9 Hz), 4.89 (d, 1 H,
H-2Ј, J = 5.9 Hz), 4.21 (m, 1 H, H-5Јa, J = 10.9 Hz), 3.91 (dd,
1 H, H-5Јb, J = 3.6, 10.9 Hz), 3.57 (dd, 1 H, H-4Ј, J = 7.6, 3.6
Hz), 3.31 (m, 2 H, 3Ј-OH, 5Ј-OH), 0.86 (s, 9 H, t-Bu), 0.18 (s,
9 H, TMS), 0.14 and Ϫ0.08 (each s, each 3 H, Me × 2); 13C
NMR (100 MHz, CDCl3) δ 166.58, 161.84, 154.96, 147.89,
132.96, 132.64, 128.68, 128.57, 127.53, 102.82, 97.75, 93.50,
82.25, 77.37, 67.21, 63.78, 54.84, 25.76, 17.89, Ϫ0.23, Ϫ4.07,
Ϫ4.63; FAB-LRMS m/z 574 (MHϩ, 54.2%); FAB-HRMS
574.2228 (MHϩ, C27H40N3O5SSi2 requires m/z 574.2227).
† CCDC reference number(s) 186372–186373. See http://www.rsc.org/
suppdata/p1/b2/b204993g/ for crystallographic files in .cif or other
electronic format.
References
1 For review article, see: M. Yokoyama, Synthesis, 2000, 12, 1637–
1655 and references therein.
2 For examples, see: (a) M. R. Dyson, P. L. Coe and R. T. Walker,
J. Med. Chem., 1991, 34, 2782–2786; (b) K. N. Tiwari, A. T.
Sfortnacy-Fowler, L. Cappellacci, W. B. Parker, W. R. Waud,
J. A. Montgomery and J. A. Secrist III, Nucleosides Nucleotides
Nucleic Acids, 2000, 19, 329–340.
3 Y. Yoshimura, K. Kitano, H. Satoh, M. Watanabe, S. Miura,
S. Sakata, T. Sasaki and A. Matsuda, J. Org. Chem., 1996, 61,
822–823.
4 (a) A. Matsuda, K. Takenuki, M. Tanaka, T. Sasaki and T. Ueda,
J. Med. Chem., 1991, 34, 812–819; (b) K. Yamagami, A. Fujii,
M. Arita, T Okumoto, S. Sakata, A. Matsuda, T. Ueda and
T. Sasaki, Cancer Res., 1991, 51, 2319–2323; (c) T. Ono, A. Fujii,
M. Hosoya, T. Okumoto, S. Sakata, A. Matsuda and T. Sasaki,
Biochem. Pharmacol., 1996, 52, 1279–1285.
5 H. Hattori, M. Tanaka, M. Fukushima, T. Sasaki and A. Matsuda,
J. Med. Chem., 1996, 39, 5005–5011.
6 (a) S. Tabata, M. Tanaka, Y. Endo, T. Obata, A. Matsuda and
T. Sasaki, Cancer Lett., 1997, 116, 225–231; (b) S. Takatori,
H. Kanda, K. Takenaka, Y. Wataya, A. Matsuda, M. Fukushima,
Y. Shimamoto, M. Tanaka and T. Sasaki, Cancer Chemother.
Pharmacol., 1999, 44, 97–104; (c) Y. Shimamoto, H. Kazuno,
Y. Murakami, A. Azuma, K. Koizumi, A. Matsuda, T. Sasaki and
M. Fukushima, Jpn. J. Cancer Res., 2002, 93, 445–452.
7 (a) T. Naka, N. Nishizono, N. Minakawa and A. Matsuda,
Tetrahedron Lett., 1999, 40, 6297–6300; (b) T. Naka, N. Minakawa,
H. Abe, D. Kaga and A. Matsuda, J. Am. Chem. Soc., 2000, 122,
7233–7243.
8 K. Nagasawa and A. Yoneta, Chem. Pharm. Bull., 1985, 33,
5048–5052.
9 A. K. Bose and B. Lal, Tetrahedron Lett., 1973, 3937–3940.
10 O. Yamazaki, H. Togo and M. Yokoyama, J. Chem. Soc.,
Perkin Trans. 1, 1999, 2891–2896.
1-(3-C-Ethynyl-4-thio-ꢀ-D-ribofuranosyl)cytosine (4)
A solution of 26 (182 mg, 0.32 mmol) in MeOH (3 mL) con-
taining ammonium fluoride (117 mg, 3.17 mmol) was heated
under reflux for 1 h. The solvent was removed in vacuo, and the
residue was dissolved in methylamine in MeOH (40%, 6 mL).
The reaction mixture was kept for 1 h at room temperature, and
the solvent was removed in vacuo. The residue was coevapor-
ated with MeOH. The residue was purified by a silica gel
column, eluted with 33% MeOH in CHCl3, to give 4 (89.0 mg,
99% as a white solid, crystallized from MeOH): mp 237 ЊC dec;
Found: C, 46.53; H, 4.71; N, 14.79. C11H13N3O4S requires C,
1
46.64; H, 4.63; N, 14.83%; H NMR (270 MHz, DMSO-d6)
2188
J. Chem. Soc., Perkin Trans. 1, 2002, 2182–2189

View Article Online
S. G. Rahim and J. Chattopadhyaya, J. Am. Chem. Soc., 1992,
11 (a) P. M. J. Jung, A. Burger and J.-F. Biellmann, Tetrahedron Lett.,
1995, 36, 1031–1034; (b) P. M. J. Jung, A. Burger and J.-F.
Biellmann, J. Org. Chem., 1997, 62, 8309–8314.
12 H. Hattori, E. Nozawa, T. Iino, Y. Yoshimura, S. Shuto,
Y. Shimamoto, M. Nomura, M. Fukushima, M. Tanaka, T. Sasaki
and A. Matsuda, J. Med. Chem., 1998, 41, 2892–2902.
114, 9936–9943; (c) J. Uenishi, K. Takahashi, M. Motoyama
and H. Akashi, Chem. Lett., 1993, 255–256; (d ) J. A. Secrist III,
K. N. Tiwari, A. T. Sfortnacy-Fowler, L. Messini, J. M. Riordan and
J. A. Montgomery, J. Med. Chem., 1998, 41, 3865–3871.
15 (a) C. Altona and M. Sundaralingam, J. Am. Chem. Soc., 1973, 95,
2333–2344; (b) D. B. Davies and S. S. Danyluk, Biochemistry, 1974,
13, 4417–4434.
13 G. H. Hakimelahi, Z. A. Proba and K. K. Ogilvie, Tetrahedron Lett.,
1981, 22, 4775–4778.
14 (a) M. Bobek, A. Bloch, R. Parthasarathy and R. L. Whistler,
J. Med. Chem., 1975, 18, 784–787; (b) L. H. Koole, J. Plavec, H. Liu,
B. R. Vincent, M. R. Dyson, P. L. Coe, R. T. Walker, G. W. Hardy,
16 C. Leydier, L. Bellon, J.-L. Barascut, J. Deydier, G. Maury,
H. Pelicano, M. A. E. Alaoui and J.-L. Imbach, Nucleosides
Nucleotides, 1994, 13, 2035–2050.
J. Chem. Soc., Perkin Trans. 1, 2002, 2182–2189
2189