A short, flexible route toward 2'-C-branched ribonucleosides

Article abstract of DOI:10.1021/jo961893+

A five-step synthesis of 2'-C-branched ribonucleosides from commercially obtained 1,3,5-tri-O-benzoyl-α-D-ribofuranose (4) is described. The free hydroxyl group of 4 was oxidized in high yield with Dess-Martin periodane reagent. The resultant 2-ketosugar was treated with MeMgBr/TiCl₄, CH₂=CHMgBr/CeCl₃, or TMSC≡CLi/CeCl₃, and in each case addition to the ketone proceeded stereoselectively to provide 2-alkylated ribofuranosides. After conversion to the corresponding tetrabenzoyl derivatives, the 2-alkylribofuranosides were coupled to nucleobases under Vorbruggen persilylation conditions, giving the β-nucleosides with high stereoselectivity. Deprotection with methanolic ammonia provided the title compounds in 17-49% overall yields from 4.

Full text of DOI:10.1021/jo961893+

1754
J . Org. Chem. 1997, 62, 1754-1759
A Sh or t, F lexible Rou te tow a r d 2′-C-Br a n ch ed Ribon u cleosid es
Rogers E. Harry-O’kuru, J ennifer M. Smith, and Michael S. Wolfe*,†
Department of Pharmaceutical Sciences, University of Tennessee, Memphis, Tennessee 38163
Received October 8, 1996X
A five-step synthesis of 2′-C-branched ribonucleosides from commercially obtained 1,3,5-tri-O-
benzoyl-R-D-ribofuranose (4) is described. The free hydroxyl group of 4 was oxidized in high yield
with Dess-Martin periodane reagent. The resultant 2-ketosugar was treated with MeMgBr/TiCl4,
CH2dCHMgBr/CeCl3, or TMSCtCLi/CeCl3, and in each case addition to the ketone proceeded
stereoselectively to provide 2-alkylated ribofuranosides. After conversion to the corresponding
tetrabenzoyl derivatives, the 2-alkylribofuranosides were coupled to nucleobases under Vorbru¨ggen
persilylation conditions, giving the â-nucleosides with high stereoselectivity. Deprotection with
methanolic ammonia provided the title compounds in 17-49% overall yields from 4.
In the past ten years, a number of 2′-C-branched
appropriately modified sugar,5 and (2) a linear approach
starting from the unmodified nucleoside.1a-g,6 The linear
approach offers a relatively rapid route to 2′â-C-branched
nucleosides; however, while this approach has provided
access to a variety of 2′â-alkyl-2′-deoxyribonucleosides
with high stereoselectivity, it has not offered the corre-
sponding 2′-alkylribonucleosides stereoselectively. Where
2′-alkylribonucleosides have been accessible using the
linear approach (specifically, via reaction of 2′-keto-
nucleoside derivatives with organometallics), the dia-
stereomeric 2′-alkylarabinonucleosides are usually the
major (and, in some cases, the exclusive) products.1a-e,6
Compared with the linear approach, the convergent
approach is potentially more flexible since a variety of
nucleobases can be coupled to the modified sugar. How-
ever, the stereoselectivity of the nucleobase coupling may
be compromised by the presence of the alkyl substituent
on the â face of the sugar.5e Reported routes to 2′-C-
branched ribonucleosides using the convergent approach5
are relatively lengthy, low yielding, or not easily ame-
nable to the incorporation of alkyl groups other than
methyl and were therefore considered suboptimal for our
purposes.
nucleosides have displayed promising anticancer1 and
antiviral2 activities as well as interesting biochemical
properties, such as DNA cleavage3 and ribonucleotide
reductase inactivation.4 We have recently initiated a
program aimed toward developing new 2′-C-branched
ribonucleosides and ribonucleotides as bioorganic tools
and as potential chemotherapeutic agents. Within this
framework, 2′-C-vinyl- and 2′-C-ethynylribonucleotides
have been designed as potential mechanism-based inac-
tivators of ribonucleotide reductases. While there are
numerous examples of 2′-C-branched 2′-deoxyribonucleo-
sides in the literature, there are surprisingly few 2′-C-
branched ribonucleosides and no 2′-C-vinyl- or 2′-C-
ethynylribonucleosides. Our initial goal was to develop
a short, flexible synthetic route toward 2′-C-branched
ribonucleosides in general and 2′-C-vinyl- or 2′-C-ethyn-
ylribonucleosides in particular.
Several synthetic methodologies have been reported
toward 2′â-C-branched nucleosides, and these can be
divided into two general strategic types: (1) a convergent
approach, where the nucleobase is glycosylated with the
A recent report7 describing the synthesis of 2-keto-
ribofuranose 1 inspired the formulation of a short,
general, convergent approach toward 2′-C-branched ri-
bonucleosides (Scheme 1). This route necessitates the
choice of organometallic reagents which selectively add
to ketone functionalities in the presence of esters. Or-
ganotitanium8 and organocerium9 reagents are known to
provide such selectivity. Given this selectivity, the
stereochemical arrangement of the adjacent substituents
should favor the addition of alkyl groups to the â face of
the ketone to provide 2-alkylsugars 2. After conversion
of the free 2-hydroxyl group to its benzoate ester, the
† Correspondence should be addressed to this author at Department
of Pharmaceutical Sciences, University of Tennessee, Room 327C, The
Faculty Building, 847 Monroe St., Memphis, TN 38163. Tel: (901)-
448-7533. Fax: (901)448-6828. E-mail: mwolfe@utmem1.utmem.edu.
X Abstract published in Advance ACS Abstracts, March 1, 1997.
(1) (a) Matsuda, A.; Takenuki, K.; Itoh, H.; Sasaki, T.; Ueda, T.
Chem. Pharm. Bull. 1987, 35, 3967. (b) Matsuda, A.; Itoh, H.;
Takenuki, K.; Sasaki, T.; Ueda, T. Chem. Pharm. Bull. 1988, 36, 945.
(c) Ueda, T.; Matsuda, A.; Yoshimura, Y.; Yakenuki, K. Nucleosides
Nucleotides 1989, 8, 743. (d) Matsuda, A.; Takenuki, K.; Sasaki, T.;
Ueda, T. J . Med. Chem. 1991, 34, 234. (e) Matsuda, A.; Nakajima, Y.;
Azuma, A.; Tanaka, M.; Sasaki, T. J . Med. Chem. 1991, 34, 2917. (f)
Yoshimura, Y.; Saitoh, K.; Ashida, N.; Sakata, S.; Matsuda, A. Bioorg.
Med. Chem. Lett. 1994, 4, 721. (g) Yoshimura, Y.; Satoh, H.; Sakata,
S.; Ashida, N.; Miyazaki, S.; Matsuda, A. Nucleosides Nucleotides 1995,
14, 427. (h) Takenuki, K.; Matsuda, A.; Ueda, T.; Sasaki, T.; Fujii, A.;
Yamagami, K. J . Med. Chem. 1988, 31, 1063. (i) Matsuda, A.;
Takenuki, K.; Tanaka, M.; Sasaki, T.; Ueda, T. J . Med. Chem. 1991,
34, 812. (j) Yamagami, K.; Fujii, A.; Arita, M.; Okumoto, T.; Sakata,
S.; Matsuda, A.; Ueda, T. Cancer Res. 1991, 51, 2319. (k) Cory, A. H.;
Samano, V.; Robins, M. J .; Cory, J . P. Biochem. Pharmacol. 1994, 47,
365.
(5) (a) Walton, E.; J enkins, S. R.; Nutt, R. F.; Zimmerman, M. Holly,
F. W. J . Am. Chem. Soc. 1966, 88, 4524. (b) J enkins, S. R.; Arison, B.;
Walton, E. J . Org. Chem. 1968, 33, 2490. (c) Walton, E.; J enkins, S.
R.; Nutt, R. F.; Holly, F. W; Nemes, M. J . Med. Chem. 1969, 12, 306.
(d) Beigelman, L. N.; Ermolinsky, B. S.; Gurskaya, G. V.; Tsapkina,
E. N.; Karpeisky, M. Y.; Mikhailov, S. N. Carbohyd. Res. 1987, 166,
219. (e) Wolf, J .; J arrige, J .-M.; Florent, J .-C.; Grierson, D. S.;
Monneret, C. Synthesis 1992, 773.
(6) (a) Yoshimura, Y.; Iino, T.; Matsuda, A. Tetrahedron Lett. 1991,
32, 6003. (b) Iino, T.; Yoshimura, Y.; Matsuda, A. Tetrahedron 1994,
50, 10397.
(7) Cook, G. P.; Greenberg, M. M. J . Org. Chem. 1994, 59, 4704.
(8) Weidman, B.; Seebach, D. Angew. Chem., Int. Ed. Engl. 1983,
22, 31.
(2) Awano, H.; Shuto, S.; Baba, M.; Kira, T.; Shigeta, S.; Matsuda,
A. Bioorg. Med. Chem. Lett. 1994, 4, 367.
(3) Matsuda, A.; Azuma, A. Nucleosides Nucleotides 1995, 14, 461.
(4) (a) Baker, C. H.; Banzon, J .; Bollinger, J . M.; Stubbe, J .; Samano,
V.; Robin, M. J .; Lippert, B.; J arvi, E.; Resvick, R. J . Med. Chem. 1991,
34, 1879. (b) Ong, S. P.; McFarlan, S. C.; Hogenkamp, H. P. C.
Biochemistry 1993, 32, 11397. (c) Bitoni, A. J .; Dumont, J . A.; Bush,
T. L.; Cashman, E. A.; Cross-Doersen, D. E.; Wright, P. S.; Matthews,
D. P.; McCarthy, J . R.; Kaplan, D. A. Cancer Res. 1994, 54, 1485. (d)
van der Donk, W. A.; Yu, G.; Silva, D. J .; Stubbe, J .; McCarthy, J . R.;
J arvi, E. T.; Matthews, D. P.; Resvick, R. J .; Wagner, E. Biochemistry
1996, 35, 8381.
(9) Imamoto, T.; Takiyama, N.; Nakamura, K.; Hatajima, T.; Ka-
miya, Y. J . Am. Chem. Soc. 1989, 111, 4392 and references therein.
S0022-3263(96)01893-2 CCC: $14.00 © 1997 American Chemical Society

Short Route to 2′-C-Branched Ribonucleosides
J . Org. Chem., Vol. 62, No. 6, 1997 1755
Sch em e 1
products 8 was about 5:3, with an overall yield of 66%.
Despite the formation of three products, the methyl group
had added with complete stereoselectivity to the â face
of the sugar; no products resulting from approach from
the R face were isolated. The organotitanium reagent
also appeared to be quite chemoselective as expected,
since no products resulting from ester cleavage were
detected either.
All three products were useful for our purposes and
converged to the tetrabenzoylated 11 upon treatment
with benzoyl chloride and DMAP in the presence of Et3N.
Regardless of whether the starting material was purified
5 or the anomeric mixture of 8, the product 11 was
isolated in about 70% yield with a â/R ratio of 4:1. Again,
the two anomers of 11 were spectroscopically identical
with previous reports.5a,b The mechanism of this reaction
involves transesterification of 5 to R-8 followed by ano-
merization of R-8 to â-8. Although these two steps are
likely reversible, â-8 is apparently the least sterically
hindered and is most rapidly benzoylated.15 This conver-
sion of 5 to â-11 has been observed before under similar
conditions (heating with benzoyl chloride in pyridine).5b
Purified â-11 was coupled with several nucleobases
under persilylation conditions.10 For the pyrimidine
bases, coupling was accomplished using SnCl4 as the
Lewis acid and refluxing in acetonitrile for 3 h. Glycos-
ylation yields were 57% and 77% for uracil and 6-aza-
uracil, respectively. Despite the presence of the 2â-
methyl substituent, only one anomer, identified as â (vide
infra), of protected nucleosides 14a and 14b was observed
in each case. A sample of 11 enriched in the R anomer
was coupled to uracil under identical conditions to give
the same product observed with purified â-11 in the same
yield and with the same stereoselectivity (i.e., the reac-
tion works through a 1,2-acyloxonium intermediate11).
Apparently, the choice of benzoyl ester protecting groups
is critical for this selectivity, since the peracetyl coun-
terpart of 11 is reported to give a 7:3 mixture of â/R
nucleoside anomers when coupled to thymine under
similar conditions.5e Coupling of â-11 with purine bases
(N6-benzoyladenine and 6-(methylthio)purine) was car-
ried out using TMSOTf as Lewis acid, refluxing for over
12 h in acetonitrile to ensure conversion of the initially
formed N-7 regioisomers to the thermodynamically fa-
vored N-9 glycosylation products.16 Again, the coupling
appeared to be completely stereoselective, forming the
protected â-nucleosides (vide infra) 14c and 14d in 86%
and 47% yields, respectively.
sugars can then be coupled to a variety of nucleobases
using Vorbru¨ggen-type persilylation conditions.10 The
steric bulkiness of the 2-benzoate group in addition to
its well-known propensity to facilitate trans approach of
the incoming nucleobase by stabilizing the sugar oxonium
intermediate11 made this the protecting group of choice.
Ammonolysis of the benzoate protecting groups would
then provide the modified nucleoside analogues 3. This
route was attractive to us because of its brevity and the
potential to install a variety of 2′ alkyl substituents as
well as a range of natural and unnatural nucleobases.
In a preliminary communication, we have reported the
synthesis of 2′-C-methylribonucleosides, using MeMgBr/
TiCl4 to form the new C-C bond.12 In this full account,
we provide a detailed description and discussion of this
work. In addition, we demonstrate that organocerium
chemistry is suitable for installing different R groups into
the sugar and illustrate the flexibility of this route by
obtaining 2′-C-vinyl- and 2′-C-ethynylribonucleosides.
Resu lts a n d Discu ssion
F or m a tion of 2′-C-Meth ylr ibon u cleosid es Usin g
MeTiCl3. Ketone 1 was formed in high yield by the
oxidation of commercially available13 1,3,5-tri-O-benzoyl-
R-D-ribofuranose (4, Scheme 2) with Dess-Martin peri-
odane reagent as previously described.7 Using this
procedure, a considerable portion of the corresponding
hydrate of 1 is also formed upon aqueous workup. In
our hands, the ratio of ketone to hydrate is generally
about 10:3. The presence of this hydrate would be
potentially detrimental to subsequent reactions with
organometallic reagents. While high-vacuum drying and
codistillation with toluene were both unsuccessful in
converting the hydrate to the ketone, stirring a CH2Cl2
solution of the ketone/hydrate mixture with a large
amount of MgSO4 overnight gave essentially pure ketone
after filtration and concentration.
Compounds 14a -d were each treated with methanolic
ammonia to provide the deprotected nucleosides 17a -
d . 2′-C-Methyluridine (17a ) and 2′-C-methyladenosine
(17c) were physically and spectroscopically identical with
previously reported data of these compounds,5 providing
unambiguous proof of their regio- and stereochemical
assignments. Evidence for the regio- and stereochemical
identity of 17b and 17d was provided by a comparison
of 13C NMR data of these two nucleosides with their
unmethylated counterparts 6-azauridine and 6-meth-
ylthiopurine riboside (Table 1) and with methylnucleo-
Ketone 1 was then treated with MeMgBr/TiCl4.14
Three products were isolated from this reaction: the
anticipated product 5 as well as its transesterified
isomers 8 as a mixture of anomers. The chromatographic
and spectral properties of these isomers matched those
in a previous report where 5 and 8 had been formed by
another route.5a,b The ratio of 5 to transesterification
(10) Vorbru¨ggen, H.; Bennua, B. Chem. Ber. 1981, 114, 1279.
(11) Paulsen, H. Adv. Carbohyd. Chem. Biochem. 1971, 26, 127.
(12) Wolfe, M. S.; Harry-O’kuru, R. E. Tetrahedron Lett. 1995, 36,
7611.
(13) Compound 4 was purchased from Pfanstiel Laboratories, Inc.,
for these studies. Recently, 4 has become unavailable from commercial
sources, but can be prepared in one step from 2,3,5-tri-O-benzoyl-1-
O-acetyl-D-ribofuranose (Pfanstiel): Brodfuehrer, P. R.; Sapino, C., J r.;
Howell, H. G. J . Org. Chem. 1985, 50, 2598.
(15) Commercially available tertiary (e.g., 1-adamantanol) and
hindered secondary alcohols (e.g., menthol) were not benzoylated at
all under the conditions employed, whereas less hindered secondary
alcohols (e.g., isomenthol) were benzoylated in good yields. These
findings, in combination with the fact that R-anomer 5 is primarily
converted to â-11, indicates that the highly hindered tertiary alcohol
5 undergoes transesterification and anomerization before benzoylation.
(16) Dudycz, L. W.; Wright, G. E. Nucleosides Nucleotides 1984, 3,
33.
(14) Reetz, M. T.; Kyung, S. H.; Hu¨llman, M. Tetrahedron 1986, 42,
2931.

1756 J . Org. Chem., Vol. 62, No. 6, 1997
Harry-O’kuru et al.
Sch em e 2a
Ta ble 1. 13C NMR Ch em ica l Sh ifts for Nu cleosid e An a logu esa
nucleobase
ribose
compound
C4
C2
C6
C5
C8
Me
C1′
C4′
C2′
C3′
C5′
substit
19.4
134.6, 119.0
78.1, 76.2
uridineb
17a
18a
166.5
166.5
166.2
165.5
158.5
158.2
149.2
149.9
148.8
146.4
158.0
157.7
157.9
152.0
152.0
151.5
152.0
149.5
149.7
152.6
153.4
152.4
151.8
166.5
166.4
166.5
142.2
141.9
141.8
141.7
102.6
102.6
102.5
102.4
137.3
137.4
119.5
119.6
132.1
130.2
96.5
89.8
92.1
91.9
91.4
90.2
92.0
88.2
91.7
88.7
91.7
90.7
92.3
92.0
84.6
82.2
82.0
82.2
84.2
83.4
86.1
83.5
86.6
82.7
84.1
82.2
81.9
74.1
79.5
82.0
79.3
73.2
79.9
73.7
79.4
74.7
79.6
74.3
81.8
79.9
69.8
72.8
71.4
74.5
70.5
74.4
70.9
72.6
71.1
72.5
69.6
71.3
74.5
61.1
60.0
59.8
59.8
61.8
62.5
61.9
60.4
62.1
60.2
61.1
60.0
59.8
19a
6-azauridineb
17b
18.6
20.7
19.3
adenosinec,d
17c
156.3
156.9
161.2
162.6
142.0
141.6
141.5
140.2
139.6
144.0
142.4
6-MTPRb,d
17d
12.1
11.8
cytidineb
18e
96.5
96.5
134.9, 118.4
77.8, 76.3
19e
a
b
All spectra obtained in D2O except where noted. Sigma Chemical Co.; MTPR (6-(methylthio)purine riboside). c Data from ref 17.
d
Spectrum obtained in DMSO-d6.
sides 17a and 17c. The nucleobase peaks of 17b and 17d
final nucleosides had been previously reported and served
to unambiguously identify the products. However, our
primary goal was to obtain previously unreported 2′-C-
vinyl- and 2′-C-ethynylribonucleosides. Although orga-
notitanium reagents are useful for adding methyl and
other alkyl groups to ketones and aldehydes, they cannot
be employed for adding vinyl and ethynyl groups (vinyl-
titanium compounds, for instance, readily undergo oxida-
tive homocoupling of the alkyl ligands in lieu of addition
to carbonyl groups8). We therefore explored the utility
of organocerium chemistry for this purpose. Organo-
cerium(III) reagents have also been reported to be
chemoselective, undergoing efficient addition to alde-
hydes and ketones in the presence of esters, even with
substrates that are susceptible to enolization.9 Reagents
of this type can be formed in situ using Grignard reagents
or alkyllithiums in the presence of cerium chloride,9 and
2′-deoxy-3′-ketonucleosides have been treated with such
reagents to form 3′-vinyl- and 3′-ethynyl-substituted
nucleosides with complete (R) diastereoselectivity.18
were virtually identical to those of their respective
unmethylated counterparts, indicating that these new
compounds were N-1 (17b) and N-9 (17d ) substituted.17
Also, the peaks belonging to the methylribose portions
of 17b and 17d were essentially identical to the meth-
ylribose moieties of the known 17a and 17c, consistent
with the stereoassignments for the methyl and nucleo-
base substituents.
F or m a tion of 2′-C-Vin yl- a n d 2′-C-Eth yn ylr ibo-
n u cleosid es Usin g Or ga n ocer iu m Ch em istr y. The
experiments described above toward 2′-C-methylribo-
nucleosides provided a convenient illustration of the
potential of our general synthetic route outlined in
Scheme 1. Several of the synthetic intermediates and
(17) (a) For 13C NMR correlation studies on N7- versus N9-
substituted purines, see: Chenon, M.-T.; Pugmire, R. J .; Grant, D. M.;
Panzica, R. P.; Townsend, L. B. J . Am. Chem. Soc. 1975, 97, 4627. (b)
For 13C NMR spectra of N1- versus N3-methyluracil derivatives, see:
Still, I. W.; Plavac, N.; McKinnon, D. M.; Chauhan, M. S. Can. J . Chem.
1978, 56, 725. (c) N3-â-D-Pyrimidineribofuranosides also display a
characteristic downfield shift of the H-1′ proton of up to 1 ppm
compared to that of the N1- nucleosides (Niedballa, U.; Vorbru¨ggen,
H. J . Org. Chem. 1974, 39, 3660). All the 1H NMR spectra of 2′-C-
alkylated pyrimidines 17-19 showed H-1′ between 5.9 and 6.1 ppm,
consistent with N1-substitution.
Treatment of ketosugar 1 with organocerium reagents
derived from vinylmagnesium bromide or lithium (tri-
(18) Bender, S. L.; Moffett, K. K. J . Org. Chem. 1992, 57, 1646.

Short Route to 2′-C-Branched Ribonucleosides
J . Org. Chem., Vol. 62, No. 6, 1997 1757
aluminum-backed plates. Flash chromatography was carried
out with 70-230 mesh silica gel 60 (EM Science). Starting
material 1,3,5-tri-O-benzoyl-R-D-ribofuranoside was purchased
from Pfanstiel Laboratories, Inc. All other chemicals/reagents
were purchased from Aldrich Chemical Co.
methylsilyl)acetylide gave a mixture of alkylated prod-
ucts, similar to that observed when MeMgBr/TiCl4 was
employed (Scheme 2).19 Partial chromatographic separa-
tion allowed for the tentative assignments of the 1,2-
addition products (6 or 7) as well as the transesterified
anomers (9 or 10). In these cases, however, several
isomers of partially debenzoylated adducts were also
detected. This was unexpected given the reported
chemoselectivity of organocerium reagents.9 While the
presence of so many products might be considered
scatological, all of these materials converged on the
tetrabenzoylated 2-alkylated sugar (12 or 13) upon
benzoyl chloride/DMAP/Et3N treatment. Furthermore,
the chromatographic cleanup of these various products
is not necessary: the two-step procedure of alkylation and
benzoylation is accomplished more easily and in good
yield (84% and 75% for 12 and 13, respectively) when
the crude intermediate alkylation products were directly
benzoylated. The 1H NMR spectra of the resultant
perbenzoylated anomers were compared with the corre-
sponding methyl-substituted sugars R-11 and â-11, and
in both cases the â-anomer was tentatively assigned as
the major product after benzoylation. For all three
perbenzoylated sugars 11-13, the H-3 peak for the major
anomer was a doublet of J ≈ 7-8 Hz, while the corre-
sponding peak for the minor anomer appeared ca. 0.3
ppm upfield as a doublet with J ≈ 2-3 Hz.
1,3,5-Tr i-O-ben zoyl-2-C-m eth yl-r-D-r ibofu r an ose (5) an d
2,3,5-Tr i-O-ben zoyl-2-C-m eth yl-r/â-D-r ibofu r an ose (8). TiCl4
(1.91 mL, 3.30 g, 17.4 mmol) was added dropwise to 55 mL of
anhydrous ether cooled to -78 °C. To the resultant yellow
etherate was slowly added 3.0 M methylmagnesium bromide
in ether (5.78 mL, 17.4 mmol); the reaction mixture was then
allowed to warm to -30 °C, whereupon a solution of 1,3,5-tri-
O-benzoyl-2-keto-R-D-ribofuranose (1)21 (2.00 g, 4.35 mmol) in
5 mL of ether was added via syringe. After 4 h at -30 to -10
°C, TLC analysis showed no 1 present, and the reaction
mixture was poured into 60 mL of water. The organic phase
was separated and the aqueous phase extracted with 3 × 50
mL of ether. The combined organic layer was washed (water),
dried (MgSO4), filtered, and concentrated to a syrup, which
was passed through silica gel (hexanes/ethyl acetate 5:2) to
give two main fractions: 0.85 g of 5 (Rf ) 0.65) and 0.51 g of
an anomeric mixture of 8 (Rf ) 0.45) for a 66% overall yield.
The spectral and/or chromatographic properties of 5 and 8
matched that of previous reports.5a,b Further purification was
not carried out, and the materials were directly benzoylated
to 11 as described below.
1,2,3,5-tetr a -O-ben zoyl-2-C-m eth yl-r-(a n d â)D-r ibofu r a -
n ose (11). To a solution of 4-(dimethylamino)pyridine (232
mg, 1.89 mmol) in 20 mL of dry CH2Cl2 and 3 mL of dry Et3N
was added benzoyl chloride (0.44 mL, 532 mg, 3.78 mmol)
followed by a solution of methylsugars 5 (0.39 g, 0.82 mmol)
and 8 (0.51 g, 1.07 mmol) in 5 mL of CH2Cl2. After 3 h at
ambient temperature, the reaction mixture was poured into
150 mL of ether and washed with 3 × 75 mL of 1 N HCl, 75
mL of saturated NaHCO3, and 75 mL of brine. The organic
phase was dried (Na2SO4), filtered, concentrated to a syrup,
and recrystallized from hexanes and ethyl acetate to give 541
mg of pure â-11. The mother liquor was concentrated and
passed through silica gel (hexane/ethyl acetate 4:1) to afford
254 mg of 11 as a mixture of anomers (R/â ratio ca 2:1
Prior to performing nucleobase couplings (uracil and
N4-acetylcytosine) with 12 and 13, we suspected that the
larger vinyl and ethynyl substituents might compromise
the high preference for forming the â-nucleoside observed
with the 2-C-methyl sugar 11. This anticipated problem
did not occur, however, and the protected â-nucleosides
15a ,e and 16a ,e were obtained in good yields (62-73%).
Deprotection provided the desired 2′-C-vinyl- and 2′-C-
ethynyluridines and cytidines 18a ,e and 19a ,e. Again,
cross-comparative analysis of the 13C NMR (Table 1)17b
and 1H NMR data17c of these new compounds with
uridine and cytidine and with methyl-substituted 17a -d
were consistent with our assigned regio- and stereochem-
istry.
In conclusion, the methodology described here allows
rapid access to 2′-C-alkylated ribonucleosides in good
overall yields. The incorporation of the unsaturated
substituents should provide useful functional handles for
obtaining other 2′-C-substituted ribonucleosides not readily
accessible by direct organometallic addition.20 We are
currently exploring this avenue as well as the conversion
of nucleosides 18 and 19 to their respective nucleotide
5′-diphosphates for in vitro testing on ribonucleotide
reductases.
1
according to H NMR). Overall yield: 795 mg (72%) and R/â
ratio ca. 1:4. â-11: mp 155-156 °C (lit.5b mp 155-156 °C). 1H
NMR (CDCl3): δ 7.15-8.15 (m, 20H), 7.09 (s, 1H), 5.97 (d, J
) 8.1 Hz, 1H), 4.80 (m, 1H), 4.70 (dd, J ) 4.0, 12.1, 1H), 4.56
(dd, J ) 4.7, 12.1, 1H), 1.97 (s, 3H). 13C NMR (CDCl3): δ 16.8,
63.5, 76.1, 78.5, 86.7, 97.8, 128.1, 128.5, 128.6, 129.5, 129.7,
129.9, 132.9, 133.4, 133.6, 133.8, 164.5, 164.7, 165.6, 166.0.
MS (FAB): m/e 459 (M+ - OBz). [R]D ) +69.7o (c ) 2.925,
CHCl3) (lit.5a +68°, c ) 1, CHCl3). Anal. Calcd for C34H28O9‚
1/4H2O: C, 69.80; H, 4.91. Found: C, 69.71; H, 4.68. r-11:
1H NMR (CDCl3, diagnostic peaks only): δ 6.85 (s, 1H), 5.70
(d, J ) 2 Hz, 1H), 2.00 (s, 3H).
P r ep a r a t ion of An h yd r ou s Cer iu m (III) Ch lor id e.
A
THF slurry of anhydrous CeCl3 was prepared using a slight
modification of a procedure described by Bender et al.18 CeCl3‚-
7H2O (99.999%) was ground to a fine powder, transferred to
3-necked flask, and heated with an oil bath to 140-160 °C
under vaccuum (0.1 mmHg) for at least 2 h. The vessel was
purged with N2 and the oil bath replaced with an ice bath.
After the CeCl3 had cooled,9 the appropriate volume of THF
(freshly distilled from benzophenone-ketyl) was slowly added
and the resultant slurry stirred overnight.
Exp er im en ta l Section
Gen er a l Meth od s. 1H and 13C NMR spectra (300 and 74.5
MHz, respectively) were recorded on
a Bruker ARX-300
1,2,3,5-Tetr a -O-ben zoyl-2-C-vin yl-r-(a n d â)D-r ibofu r a -
n ose (12). To a -78 °C suspension of CeCl3 (17.2 g of
heptahydrate, 46.2 mmol, dried as described above) in THF
(200 mL) was added vinylmagnesium bromide (1 M in THF,
42.6 mL, 42.6 mmol) over 20 min. The mixture was stirred
for 1.5 h, whereupon a solution of dried 121 (4.87 g, 10.6 mmol)
in 100 mL of THF was transferred in via cannula over 15-20
min. After an additional 30 min, TLC showed no detectable
1, whereupon the dry ice/acetone bath was removed and 100
mL of saturated NH4Cl was poured into the reaction mixture.
After being allowed to warm to ambient temperature, the
instrument. 13C NMR data for compounds 17-19 are listed
in Table 1. Mass spectra were obtained on an AutoSpecQ
hybrid (E1BE2qQ) mass spectrometer (VG Fisons, Altrincham,
U.K.) using 3-nitrobenzyl alcohol as matrix. Thin-layer chro-
matography was run on Whatman 250 µm AL SIL G/UV
(19) Ketosugar 1 was also treated with MeMgBr/CeCl3 to give
results similar to those when MeMgBr/TiCl4 was added to this
ketone: a mixture of 5 (major product) and the anomers 8 (minor
products).
(20) Functionalized 2′-CR-branched 2′-deoxyuridines have recently
been reported via the corresponding 2′-CR-allylic uridine: Lawrence,
A. J .; Pavey, J . B. J .; Cosstick, R.; O’Neil, I. A. J . Org. Chem. 1996,
61, 9213.
(21) Obtained according to ref 7 and dried with MgSO4 in CH2Cl2
overnight.

1758 J . Org. Chem., Vol. 62, No. 6, 1997
Harry-O’kuru et al.
mixture was filtered through Celite, washing the filtered solid
with 3 × 100 mL of ether and 100 mL of saturated NH4Cl.
The organic layer of the filtrate was further extracted with 3
× 100 mL of saturated NH4Cl, and the combined aqueous
phase was washed with 200 mL of ether. The combined
organic phases were then dried (Na2SO4) and concentrated to
provide 5.7 g of a viscous yellow oil. This material was taken
up in 250 mL of dry CH2Cl2, whereupon 2.50 g (20.5 mmol) of
(dimethylamino)pyridine, 4.60 mL (39.6 mmol) of benzoyl
chloride, and 25 mL of distilled Et3N were added. After 4 h,
the reaction mixture was worked up as described for 11 to yield
5.27 g of 12 (84% overall from 1) as a mixture of anomers (R/â
ratio ca. 1:4). 1H NMR (CDCl3): δ 4.58 (dd, J ) 12.1, 4.7 Hz,
1H, â), 4.70-4.95 (m, 3H, R and â), 5.42 (d, J ) 11.3 Hz, 1H,
â), 5.48 (d, J ) 17.7, 1H, â), 5.54 (d, J ) 11.0 Hz, 0.2H, R),
5.80 (d, J ) 17.6 Hz, 0.2H, R), 5.87 (d, J ) 3.0 Hz, 0.2H, R),
6.29 (d, J ) 8.2 Hz, 1H, â), 6.40 (dd, J ) 17.6, 11.2, 0.2 H, R),
6.49 (dd, J ) 17.7, 11.3 Hz, 1H, â), 7.15-8.30 (m, 27H, R and
â). 13C NMR (CDCl3; â peaks only): δ 63.8, 73.8, 78.5, 87.2,
97.2, 119.2, 128.1, 128.6, 129.6, 129.8, 129.9, 130.0, 132.6,
133.6, 133.6, 133.7, 164.2, 164.4, 165.3, 166.0. MS (FAB): m/e
471 (M+ - OBz). Anal. Calcd for C35H28O9: C, 70.94; H, 4.76.
Found: C, 70.65; H, 4.96.
1:4. r-13. 1H NMR (CDCl3): δ 0.26 (s, 9H), 4.75-4.95 (m,
3H), 6.01 (d, J ) 2.0 Hz), 7.14 (s, 1H), 7.20-7.65 (m, 12H),
7.45-8.20 (m, 8H). 13C NMR (CDCl3): δ -0.4, 65.0, 75.9, 76.2,
83.1, 96.0, 98.6, 98.8, 128.3, 128.3, 128.4, 128.4, 128.8, 129.0,
129.4, 129.6, 129.8, 129.9, 129.9, 133.1, 133.4, 133.5, 133.6,
163.5, 164.1, 165.0, 166.1. MS (FAB): m/e 663 (M + 1), 541
(M+ - OBz). [R]D ) +18° (c 21.75, CHCl3). Anal. Calcd for
C38H34O9Si: C, 68.87; H, 5.17. Found: C, 68.82; H, 5.31. â-13:
mp 122-123 °C. 1H NMR (CDCl3): δ 0.11 (s, 9H), 4.65 (m,
1H), 4.81 (m, 2H), 6.42 (d, J ) 7.3 Hz, 1H), 7.04 (s, 1H), 7.16
(m, 2H), 7.35-7.70 (m, 10H), 7.97 (m, 2H), 8.17 (m, 6H). 13C
NMR (CDCl3): δ -0.6, 63.5, 75.8, 79.1, 79.5, 96.0, 96.2, 97.3,
128.0, 128.4, 128.5, 128.5, 128.8, 129.1, 129.2, 129.4, 129.5,
129.8, 129.8, 130.0, 132.8, 133.5, 133.5, 133.6, 163.5, 164.2,
164.9, 165.9. MS (FAB): m/e 663 (M + 1), 459 (M+ - OBz).
[R]D ) +6.4° (c 5.365, CHCl3). Anal. Calcd for C38H34O9Si:
C, 68.87; H, 5.17. Found: C, 68.63; H, 5.20.
2′,3′,5′-Tr i-O-ben zoyl-2′-C-m eth ylu r id in e (14a ). To a
suspension of uracil (176 mg, 1.57 mmol) in 5 mL of distilled
acetonitrile was added N,O-bis(trimethylsilyl)acetamide (0.775
mL, 638 mg, 3.14 mmol), and the resultant solution was
brought to reflux for 30 min. The solution was allowed to cool
to ambient temperature, whereupon a solution of â-11 (430
mg, 0.74 mmol) in 5 mL of acetonitrile was added. SnCl4
(0.300 mL, 668 mg, 2.56 mmol) was added slowly dropwise,
and the reaction mixture was brought to reflux for 3 h. The
cooled reaction mixture was then taken up into 200 mL of ethyl
acetate and washed with 3 × 125 mL of saturated NaHCO3
and 125 mL of brine. The organic layer was dried (Na2SO4),
filtered, and concentrated, and the resultant syrup was passed
through silica gel (ethyl acetate/hexanes 1:1) to afford 239 mg
(57%) of 14a as a white foam. A small amount was recrystal-
lized from ethyl acetate and hexanes for analytical purposes:
1,3,5-Tr i-O-b en zoyl-2-C-[(t r im et h ylsilyl)et h yn yl]-r-D-
r ibofu r a n ose (7). To a -78 °C solution of (trimethylsilyl)-
acetylene (0.72 mL, 0.50 g, 5.1 mmol) in 25 mL of THF was
added n-butyllithium (1.6 M in hexanes, 3.12 mL, 5.0 mmol)
dropwise over 5 min. After 30 min, the solution was trans-
ferred via cannula to a -78 °C suspension of CeCl3 (2.04 g of
heptahydrate, 5.48 mmol, dried as described above) in 10 mL
of THF dropwise over 10-15 min. After 1.5 h, a solution of
121 (530 mg, 1.15 mmol) in 6 mL of THF was added over 5
min. After another 2 h, there was no detectable 1 by TLC
(hexane:EtOAc 7:1). The reaction mixture was quenched and
worked up as described for above for 12. The crude residue
was then passed through silica gel (hexane:EtOAc 85:15, then
75:25, then 65:35). The first material off the column was 7
(377 mg, 59%), which crystallized from hexane:EtOAc:CH2Cl2.
Other fractions from the column were determined by 1H NMR
to contain mixtures of other alkylated products (154 mg
total): that is, trimethylsilyl protons, sugar ring protons, and
benzoyl protons were all present. These other products were
combined and benzoylated as described below. 7: mp 136-
138 °C. 1H NMR (CDCl3): δ 0.26 (s, 9H), 3.34 (br s, 1H), 4.70
(m, 2H), 4.88 (m, 1H), 5.58 (d, J ) 2.3 Hz, 1H), 6.67 (s, 1H),
7.48 (m, 6H), 7.60 (m, 3H), 8.12 (m, 6H). 13C NMR (CDCl3):
δ 0.5, 64.9, 74.0, 76.6, 82.9, 94.5, 99.0, 101.8, 128.3, 129.3,
mp 201-202 °C (lit.5c
mp 200-201 °C). 1H NMR (CDCl3): δ
9.12 (br s, 1H), 8.09 (m, 4H), 7.90 (m, 2H), 7.38-7.66 (m, 8H),
7.22-7.36 (m, 2H), 6.55 (s, 1H), 5.79 (d, J ) 5.3 Hz, 1H), 5.74
(d, J ) 8.2 Hz, 1H), 4.77-4.97 (m, 2H), 4.60-4.71 (m, 1H),
1.78 (s, 3H). 13C NMR (CDCl3): δ 18.0, 63.4, 75.5, 80.4, 84.1,
89.4, 102.4, 128.4, 128.6, 129.5, 129.7, 129.8, 130.0, 133.5,
133.6, 133.7, 140.8, 149.9, 162.8, 165.2, 165.3, 166.2. MS
(FAB): m/e 571 (M + 1). [R]D ) -23° (c 0.315, CHCl3) (lit.5c
-23°, c 1, CHCl3). Anal. Calcd for C31H26N2O9: C, 65.26; H,
4.59; N, 4.91. Found: C, 65.01; H, 4.47; N, 4.81.
2′,3′,5′-Tr i-O-ben zoyl-2′-C-vin ylu r id in e (15a ) was pre-
pared according to the procedure described above for the
synthesis of 14a with (a) 110 mg of uracil (0.912 mmol), 0.460
mL of N,O-bis(trimethylsilyl)acetamide (379 mg, 1.86 mmol)
in 4 mL of acetonitrile; (b) 270 mg of 12 (0.456 mmol) in 6 mL
of acetonitrile; and (c) 0.170 mL of SnCl4 (378 mg, 1.45 mmol).
The reaction mixture was heated to reflux for 3 h, cooled,
diluted with 50 mL of EtOAc, and poured into 80 mL of
saturated NaHCO3. The mixture was stirred until efferves-
cence ceased and then filtered through Celite, washing the
filter cake with 3 × 50 mL of ethyl acetate. The organic layer
was separated and the aqueous phase extracted with 2 × 30
mL of ethyl acetate. The combined organic layers were dried
(Na2SO4), filtered, and concentrated, and the residue was
passed through silica gel (hexanes/EtOAc 1:1) to provide 164
mg (62%) of 15a : mp 115-118 °C. 1H NMR (CDCl3): δ 9.6
(br s, 1H), 8.11 (m, 4H), 7.86 (m, 2H), 7.20-7.70 (m, 10H),
6.66 (s, 1H), 6.13 (dd, J ) 17.5, 11.1 Hz, 1H), 6.06 (d, J ) 5.2,
1H), 5.64 (dd, J ) 8.2, 2.0 Hz, 1H), 5.45 (two d, J ) 17.6, 11.1
Hz, 2H), 4.95 (dd, J ) 12.3, 3.2 Hz, 1H), 4.82 (dd, J ) 12.3,
5.7 Hz, 1H), 4.66 (m, 1H). 13C NMR (CDCl3): δ 63.5, 73.8,
81.2, 85.8, 89.6, 102.3, 120.9, 128.8, 128.9, 129.0, 129.1, 130.1,
130.2, 130.5, 131.0, 133.9, 134.0, 134.1, 141.4, 150.1, 162.7,
129.8, 129.9, 133.1, 133.6, 133.7, 164.9, 165.4, 166.0. [R]D
)
+23° (c 0.665, CHCl3). MS (FAB): m/e 559 (M + 1). Anal.
Calcd for C31H30O8Si: C, 66.65; H, 5.41. Found: C, 66.57; H,
5.49.
1,2,3,5-Tet r a -O-b en zoyl-2-C-t r im et h ylsilyet h yn yl-r/â-
D-r ibofu r a n ose (13). Meth od A. To a solution of 7 (286 mg,
0.51 mmol) in 6 mL of dry CH2Cl2 was added 61 mg of
4-(dimethyamino)pyridine (0.50 mmol), 0.115 mL of benzoyl
chloride (140 mg, 1.0 mmol), and 0.60 mL of Et3N. After 3 h,
the reaction mixture was worked up as described above for
11 and the residue passed through silica gel (hexane:CH2Cl2:
EtOAc 19:1:1, then 12:1:1, then 8:1:1). Using this chromato-
graphic procedure, R-13 and â-13 were partially separated for
analysis. â-13 was recrystallized from hexane:EtOAc:CH2Cl2
to give white needles. Total yield after chromatography: 339
mg (100%). Ratio of R to â: 1:4.
The mixture of other products isolated during the formation
of 7 (154 mg) was likewise benzoylated using the same
amounts of reagents. After workup and chromatography
(hexane:EtOAc:CH2Cl2), 170 mg of 13 was isolated as a
mixture of anomers (R/â 1:4). Since the starting material for
this benzoylation reaction was a mixture of compounds that
were not all characterized, a yield for this process could not
be calculated; however, the overall yield from 1 was 82%.
Meth od B. After the organocerium reaction (from 5.3 g,
11.5 mmol of 3), the crude alkylation products were benzoy-
lated together directly as in method A, except 4 equiv of
benzoyl chloride was used. After chromatography, 5.70 g of
13 (75% overall yield from 3) was isolated. Ratio of R to â:
164.9, 165.5, 166.6. MS (FAB): m/e 583 (M + 1). [R]D
) -21°
(c 0.230, CHCl3). Anal. Calcd for C32H26N2O9: C, 65.98; H,
4.50; N, 4.81. Found: C, 65.98; H, 4.45; N, 4.79.
N4-Acetyl-2′,3′,5′-tr i-O-ben zoyl-2′-C-vin ylcytid in e (15e)
was prepared according to the procedure described above for
the synthesis of 14a with (a) 260 mg of N4-acetylcytosine (1.70
mmol), 0.83 mL of N,O-bis(trimethylsilyl)acetamide (683 mg,
3.36 mmol) in 4 mL of acetonitrile; (b) 440 mg of 12 (0.743
mmol) in 6 mL of acetonitrile; and (c) 0.400 mL of SnCl4 (890
mg, 3.42 mmol). The reaction mixture was heated to reflux

Short Route to 2′-C-Branched Ribonucleosides
J . Org. Chem., Vol. 62, No. 6, 1997 1759
for 15 min and then cooled to ambient temperature. After 3
h, the mixture was worked up as described above for 15a , and
the resulting residue was passed through silica gel (hexanes/
EtOAc/EtOH 5:4:1) to provide 300 mg (65%) of 15e: mp 128-
129 °C. 1H NMR (CDCl3): δ 10.63 (br s, 1H), 8.10 (m, 4H),
7.82 (m, 3H), 7.10-7.65 (m, 10H), 6.85 (br s, 1H), 6.06 (m, 2H),
5.33 (two d, J ) 17.4, 11.4 Hz, 2H), 4.90 (m, 2H), 4.73 (m,
1H), 2.30 (s, 3H). 13C NMR (CDCl3): δ 25.3, 63.6, 74.1, 81.4,
86.1, 90.0 (br), 96.9, 120.4, 128.7, 128.8, 128.9, 129.0, 130.1,
130.2, 130.5, 133.9, 134.0, 134.1, 146.1, 155.1, 163.8, 164.9,
165.4, 166.6, 171.8. MS (FAB): m/e 624 (M + 1). [R]D ) -35°
(c 1.05, CHCl3). Anal. Calcd for C34H29N3O9: C, 65.48; H, 4.69;
N, 6.74. Found: C, 65.50; H, 4.77; N, 6.86.
2′,3′,5′-Tr i-O-ben zoyl-2′-C-[(tr im eth ylsilyl)eth yn yl]u r i-
d in e (16a ) was prepared according to the procedure described
above for the synthesis of 14a with (a) 319 mg of uracil (2.85
mmol), 1.40 mL of N,O-bis(trimethylsilyl)acetamide (1.15 g,
5.66 mmol) in 5 mL of acetonitrile; (b) 925 g of 13 (1.39 mmol)
in 5 mL of acetonitrile; and (c) 0.700 mL of SnCl4 (1.56 g, 5.98
mmol). The reaction mixture was quenched and extracted as
described above for 15a . The combined organic layers were
further washed with 3 × 50 mL of saturated NaHCO3 and 50
mL of brine, dried (Na2SO4), filtered, and concentrated, and
the residue was passed through silica gel (hexanes/EtOAc 3:2
followed by 1:1) to provide 562 mg (62%) of 16a : mp 160-161
°C. 1H NMR (CDCl3): δ 8.70 (br s, 1H), 8.13 (m, 4H), 7.92 (d,
J ) 8.3 Hz, 1H), 7.85 (m, 2H), 7.40-7.65 (m, 8H), 7.27 (m,
2H), 6.73 (s, 1H), 5.99 (d, J ) 2.1 Hz, 1H), 5.80 (d, J ) 8.2 Hz,
1H), 5.15 (dd, J ) 11.6, 8.7 Hz, 1H), 4.75 (dd, J ) 11.6, 4.0
Hz, 1H), 4.67 (m, 1H), 0.23 (s, 9H). 13C NMR (CDCl3): δ -0.3,
64.3, 76.3, 77.9, 82.6, 87.6, 96.8, 100.7, 101.7, 128.7, 128.8,
128.9, 129.0, 130.0, 130.2, 130.3, 130.6, 133.7, 134.1, 141.5,
150.4, 163.1, 164.1, 165.1, 166.5. MS (FAB): m/e 653 (M +
1H), 3.93 (m, 2H), 3.76 (m, 2H), 1.11 (s, 3H). MS (FAB): m/e
259 (M + 1). [R]D ) +84.9° (c 0.695, H2O) (lit.5d +82°, c 0.7,
H2O). Anal. Calcd for C10H14N2O6‚2/3H2O: C, 44.45; H, 5.72;
N, 10.37. Found: C, 44.74; H, 5.84; N, 10.10.
2′-C-Vin ylu r id in e (18a ). Deprotection of 164 mg of 15a
(0.28 mmol) in methanolic ammonia for 24 h was followed by
removal of solvent in vacuo. The residue was redissolved in
water (20 mL) and washed with 3 × 20 mL of CCl4. The
aqueous layer was concentrated and the residue passed
through silica gel (EtOAc/MeOH 9:1 followed by 4:1) to give
70 mg (90%) of 18a : mp 173-174 °C. 1H NMR (D2O): δ 7.88
(d, J ) 8.1 Hz, 1H), 5.90 (s, 1H), 5.81 (d, J ) 8.1 Hz, 1H), 5.64
(dd, J ) 17.3, 10.8 Hz, 1H), 5.32 (two d, J ) 17.3, 10.6 Hz,
2H), 4.15 (d, J ) 9.4 Hz, 1H), 4.00 (m, 2H), 3.83 (dd, J ) 13.1,
3.7 Hz, 1H). MS (FAB): m/e 271 (M + 1). [R]D ) +82° (c 0.505,
MeOH). Anal. Calcd for C11H14N2O6: C, 48.89; H, 5.22; N,
10.37. Found: C, 48.73; H, 5.22; N, 10.28.
2′-C-Vin ylcytid in e (18e). Deprotection of 577 mg of 15e
(0.93 mmol) and workup was as described above for 18a . The
resultant residue was passed through silica gel (EtOAc/MeOH
4:1 followed by 2:1) to give 243 mg (83%) of 18e: mp 202-204
°C. 1H NMR (D2O): δ 7.80 (d, J ) 7.6 Hz, 1H), 5.95 (m, 2H),
5.53 (dd, J ) 17.3, 10.8 Hz, 1H), 5.28 (d, J ) 16.7, 1H), 5.18
(d, J ) 10.9 Hz, 1H), 4.11 (d, J ) 9.4 Hz, 1H), 4.00 (m, 2H),
3.80 (dd, J ) 13.0, 3.7 Hz, 1H). 13C NMR (D2O): δ 60.0, 71.4,
81.8, 82.2, 92.3, 96.5, 118.5, 134.9, 141.6, 157.7, 166.3. MS
(FAB): m/e 270 (M + 1). [R]D ) +94° (c 0.48, MeOH). Anal.
Calcd for C11H15N3O5‚H2O C, 45.99; H, 5.96; N, 14.63.
Found: C, 46.28; H, 5.86; N, 14.49.
2′-C-Eth yn ylu r id in e (19a ). Deprotection of 511 mg of 16a
(0.78 mmol) and workup was as described above for 18a , except
the reaction ran for 2 days. The resultant residue was passed
through silica gel (EtOAc followed by EtOAc/MeOH 9:1) to give
210 mg (91%) of 19a : mp 161-162 °C. 1H NMR (D2O): δ 7.95
(d, J ) 8.1 Hz, 1H), 6.04 (s, 1H), 5.89 (d, J ) 8.1 Hz, 1H), 4.22
(d, J ) 9.1 Hz, 1H), 4.02 (m, 2H), 3.83 (dd, J ) 13.4, 4.4 Hz,
1H), 3.03 (s, 1H, slowly exchanges). MS (FAB): m/e 269 (M
+ 1). [R]D ) +124° (c 0.505, MeOH). Anal. Calcd for
C11H12N2O6: C, 49.26; H, 4.51; N, 10.44. Found: C, 49.56; H,
4.52; N, 10.35.
2′-C-Eth yn ylcytid in e (19e). Deprotection of 670 mg of 16e
(0.97 mmol) and workup was as described above for 18a , except
the reaction ran for 2 days. The resultant residue was passed
through silica gel (EtOAc/MeOH 5:2) to give 234 mg (91%) of
19e. Recrystallization from EtOH provided an analytically
pure sample: mp 233-234 °C dec. 1H NMR (D2O): δ 2.85 (s,
1H, slowly exchanges), 3.75 (dd, J ) 3.9, 13.0 Hz, 1H), 3.95
(m, 2H), 4.11 (d, J ) 9.2 Hz, 1H), 5.96 (d, J ) 7.6 Hz, 1H),
6.01 (s, 1H), 7.78 (d, J ) 7.6 Hz, 1H). MS (FAB): m/e 268 (M
+ 1). [R]D ) +162° (c 0.25, MeOH). Anal. Calcd for
C11H13N3O5: C, 49.44; H, 4.90; N, 15.72. Found: C, 49.17; H,
4.81; N, 15.36.
1). [R]D
) +11° (c 0.96, CHCl3). Anal. Calcd for
C35H32N2O9Si: C, 64.40; H, 4.94; N, 4.29. Found: C, 64.29;
H, 4.98; N, 4.35.
N4-Acet yl-2′,3′,5′-t r i-O-b en zoyl-2′-C-[(t r im et h ylsilyl)-
eth yn yl]cytid in e (16e) was prepared according to the pro-
cedure described above for the synthesis of 14a with (a) 933
mg of N4-acetylcytosine (6.10 mmol), 3.00 mL of N,O-bis-
(trimethylsilyl)acetamide (2.47 g, 12.1 mmol) in 6 mL of
acetonitrile; (b) 1.88 g of 13 (2.84 mmol) in 10 mL of aceto-
nitrile; and (c) 1.43 mL of SnCl4 (3.18 g, 12.1 mmol). The
reaction mixture was heated to reflux for 3.5 h, whereupon
the mixture was worked up as described above for 16a and
the residue passed through silica gel (hexanes/EtOAc 3:2
followed by 1:1) to provide 1.43 g (73%) of 16e: mp 130-131
°C. 1H NMR (CDCl3): δ 8.77 (br s, 1H), 8.26 (d, J ) 7.6 Hz,
1H), 8.13 (m, 4H), 7.85 (m, 2H), 7.65-7.38 (m, 9H), 7.24 (d, J
) 7.6 Hz, 1H), 7.00 (s, 1H), 6.01 (d, J ) 1.7 Hz, 1H), 5.21 (dd,
J ) 11.4, 8.7 Hz, 1H), 4.80-4.64 (m, 2H), 2.27 (s, 3H), 0.23 (s,
9H). 13C NMR (CDCl3): δ -0.3, 25.4, 64.4, 76.5, 78.7, 82.8,
87.9, 96.6, 96.9, 100.9, 128.8, 128.9, 129.0, 130.0, 130.2, 130.3,
130.7, 133.7, 134.0, 134.1, 146.4, 155.2, 163.8, 164.3, 165.2,
166.5, 171.6. MS (FAB): m/e 694 (M + 1). [R]D ) -12.1° (c
0.54, CHCl3). Anal. Calcd for C37H35N3O9Si: C, 64.06; H, 5.08;
N, 6.06. Found: C, 63.75; H, 4.97; N, 5.70.
Ack n ow led gm en t. This project was supported by
Grant IN-176-D from the American Cancer Society,
through the American Association of Colleges of Phar-
macy Grant Program for New Investigators, and through
USPHS Grant CA-70901-01.
2′-C-Meth ylu r id in e (17a ). An ice-cold solution of 199 mg
(0.349 mmol) of 14a in 40 mL of methanol was saturated with
ammonia gas and the vessel stoppered. After 2 days, the
solvent was removed and the residue was taken up in 15 mL
of water and washed with 3 × 15 mL of CH2Cl2 and 15 mL of
ether. The aqueous phase was concentrated, residual water
being removed via ethanol azeotrope, and the resulting mate-
rial was passed through silica gel (ethyl acetate/methanol 9:1
followed by 4:1) to afford 78 mg (87%) of 17a as a white solid:
mp 101-103 °C (lit.5d softened at 101 °C). 1H NMR (CDCl3):
δ 7.82 (d, J ) 8.0 Hz, 1H), 5.92 (s, 1H), 5.81 (d, J ) 8.1 Hz,
Su p p or tin g In for m a tion Ava ila ble: Full spectral and
analytical data for compounds 14b-d and 17b-d (3 pages).
This material is contained in libraries on microfiche, im-
mediately follows this article in the microfilm version of the
journal, and can be ordered from the ACS; see any current
masthead page for ordering information.
J O961893+