Disiloxane-protected 2-deoxyribonolactone as an efficient precursor to 1,2-dideoxy-1-β-aryl-D-ribofuranoses

Article abstract of DOI:10.1021/ol990813w

(matric presented) Aryl C-nucleosides are analogues of natural nucleosides where the bases have been replaced with aromatic moieties. Work herein describes the highly stereoselective syntheses of non-hydrogen-bonding carbocyclic derivatives using a disiloxane-protected 2-deoxy-D-ribono-1,4-lactone as a stable and readily accessible starting material. Unlike the bis(TBDMS)-protected congener, this compound enables the use of sterically congested ortho-substituted aryllithium reagents in the initial addition reaction.

Full text of DOI:10.1021/ol990813w

ORGANIC
LETTERS
1999
Vol. 1, No. 8
1173-1175
Disiloxane-Protected
2-Deoxyribonolactone as an Efficient
Precursor to
1,2-Dideoxy-1-â-aryl-D-ribofuranoses
Uthai Wichai and Stephen A. Woski*
Department of Chemistry and Coalition for Biomolecular Products,
The UniVersity of Alabama, Tuscaloosa, Alabama 35487-0336
swoski@bama.ua.edu
Received July 14, 1999
ABSTRACT
Aryl C-nucleosides are analogues of natural nucleosides where the bases have been replaced with aromatic moieties. Work herein describes
the highly stereoselective syntheses of non-hydrogen-bonding carbocyclic derivatives using a disiloxane-protected 2-deoxy-D-ribono-1,4-
lactone as a stable and readily accessible starting material. Unlike the bis(TBDMS)-protected congener, this compound enables the use of
sterically congested ortho-substituted aryllithium reagents in the initial addition reaction.
There has been much interest in the synthesis of nucleoside
derivatives bearing carbocyclic aromatic moieties in the place
of the nucleobases. Aryl C-nucleosides maintain the ability
for aromatic stacking while relinquishing their hydrogen-
bonding potentials. Due to these characteristics, such mol-
ecules have been studied as potential universal bases1 and
as non-hydrogen-bonding isosteres of natural bases.2 More-
over, there is much potential for the use of such nucleoside
replacements in the design of novel base pairs involving
hydrophobic interactions and shape complementarity.3
mimics the anomeric stereochemistry of natural nucleosides.
The most common method for the synthesis of these
compounds involves the reaction of diarylcadmium reagents
with 1,2-dideoxy-3,5-di-O-p-toluoyl-R-1-chloro-D-ribofura-
nose.4 Unexpectedly, these substitution reactions do not
proceed with inversion but yield the R-anomers as the major
products in moderate yields.4 The R-anomers can be equili-
brated under acidic conditions to mixtures favoring the
â-anomers, providing moderate yields of the desired com-
pounds.4
The lack of stereochemical control in the addition reaction
and the limited stability of the key chlorofuranoside5
intermediate prompted this laboratory to explore alternative
routes to aryl C-nucleosides. Recently, we reported a method
for the synthesis of 1,2-dideoxy-â-1-phenyl-D-ribofuranose
Despite the relatively straightforward structures of these
compounds, efficient methods for the synthesis of simple
aryl C-nucleosides are scarce. The key synthetic issue is the
incorporation of the aryl moiety in the â-configuration, which
(1) Millican, T. A.; Mock, G. A.; Chauncey, M. A.; Patel, T. P.; Eaton,
M. A. W.; Gunning, J.; Cutbush, S. D.; Neidle, S.; Mann, J. Nucleic Acids
Res. 1984, 12, 7435.
(2) (a) Schweitzer, B. A.; Kool, E. T. J. Org. Chem. 1994, 59, 7238. (b)
Schweitzer, B. A.; Kool, E. T. J. Am. Chem. Soc. 1995, 117, 1863.
(3) Mattray, T. J.; Kool, E. T. J. Am. Chem. Soc. 1998, 120, 6191.
(4) (a) Ren, R. X.-F.; Chaudhuri, N. C.; Paris, P. L.; Rumney, S.; Kool,
E. T. J. Am. Chem. Soc. 1996, 118, 7671. (b) Guckian, K. M.; Schweitzer,
B. A.; Ren, R. X.-F.; Sheils, C. J.; Paris, P. L.; Tahmassebi, D. C.; Kool,
E. T. J. Am. Chem. Soc. 1996, 118, 8182. (c) Chaudhuri, N. C.; Ren, R.
X.-F.; Kool, E. T. Synlett 1997, 341.
(5) Hoffer, M. Chem. Ber. 1960, 93, 2777.
10.1021/ol990813w CCC: $18.00 © 1999 American Chemical Society
Published on Web 09/17/1999

(1a) based upon the addition of phenyllithium to a protected
2-deoxy-D-ribono-1,4-lactone followed by stereoselective
reduction of the resulting hemiketal.6 This result demon-
strated that approaches used for syntheses of aryl ribofura-
noses7 could be applied to the efficient and stereoselective
synthesis of a â-aryl 2-deoxyribofuranose. Significantly, this
method utilizes an easily prepared, shelf-stable precursor.
The modification of this methodology to produce a general
synthesis of other aryl C-nucleosides is reported in this Letter.
Initial efforts focused on the addition of several aryllithium
reagents to the 3,5-di-O-TBDMS ether of 2-deoxy-D-ribono-
1,4-lactone (2a) used previously (Scheme 1).6 Unexpectedly,
Scheme 2
Scheme 1
this reaction proved to be very sensitive to substitution on
the aryllithium reagent. For example, reactions of 2a with
4-tolyllithium, 3-tolyllithium, or 2-naphthyllithium followed
by treatment with Et3SiH/BF3‚OEt2 produced significantly
diminished yields of the C-nucleosides (10, 8, and 16%,
respectively). Aryllithiums with substitution adjacent to the
carbanion (2-tolyllithium and 1-naphthyllithium) produced
none of the desired product. Because the desired C-
nucleosides and unreacted starting material were the only
products isolated, it appeared that the addition of the
aryllithium reagent was the problematic step. It was postu-
lated that the O-protecting groups hindered the approach of
the aryllithium reagents toward the lactone carbonyl carbon.
Although these groups appear remote to the reactive site,
the trajectory of the aryllithium reagent must bring it directly
over the furanose ring during the course of the reaction. It
appears likely that the TBDMS groups are large enough to
sterically hinder this approach. One possible means to reduce
the steric influence of these groups is to restrict their motion
by imposing a cyclic structure. The bifunctional disiloxane
protecting group introduced by Markiewicz8 is suitable for
this purpose, and the cyclic 5′,3′-disiloxane derivatives of
nucleosides are well-known compounds. Synthesis of the
required deoxyribonolactone disiloxane (2b) was accom-
plished in two steps (Scheme 2). First, oxidation of 2-deoxy-
D-ribose by aqueous bromine generates the corresponding
2-deoxy-D-ribono-1,4-lactone. Reaction of this crude product
with 1,3-dichloro-1,1,3,3-tetraisopropyldisiloxane and imi-
dazole produces the protected 2-deoxy-D-ribono-1,4-lactone
in excellent overall yields (g85%) on a multigram scale.
Conversion of the 2-deoxy-D-ribono-1,4-lactone to the aryl
C-nucleosides was attempted using the procedure previously
employed by this laboratory.6 Thus, aryllithium reagents were
added to a solution of lactone 2b at -78 °C (Scheme 2).
The reaction was maintained at low temperature for an hour,
when it was quenched. The crude products, which presum-
ably include both diastereomeric hemiketals and the keto
alcohols, were treated at -78 °C with Et3SiH in the presence
of a strong Lewis acid. This produces the corresponding
C-nucleosides 3a-g in higher yields than were observed with
the TBDMS-protected starting material. Moreover, the reac-
tion proved less sensitive to the nature of the aryllithium
reagent, producing moderate yields of products even with
sterically hindered reagents such as 1-naphthyllithium and
2-tolyllithium (Table 1).
One drawback to the use of the disiloxane protecting group
is the observed formation of some of the R-diastereomer.
Table 1. Yields and â:R Ratios of 3a-g from the Reaction of
ArLi with 2b Followed by Reduction with Et3SiH/BF3‚OEt2
(6) Wichai, U.; Woski, S. A. Bioorg. Med. Chem. Lett. 1998, 8, 3465.
(7) (a) Matulic-Adamic, J.; Beigelman, L.; Portmann, S.; Egli, M.;
Usman, N. J. Org. Chem. 1996, 61, 3909. (b) Matulic-Adamic, J.;
Beigelman, L. Tetrahedron Lett. 1997, 38, 203. (c) Matulic-Adamic, J.;
Beigelman, L. Tetrahedron Lett. 1997, 38, 1669.
(8) Markiewicz, W. T. J. Chem. Res, Synop. 1979, 24.
1174
Org. Lett., Vol. 1, No. 8, 1999

The â/R ratios (determined by 1H NMR spectroscopy) were
generally about 10:1 (Table 1), and the anomers can be
separated by flash column chromatography. The configura-
tions of the â-anomers of 3a-g were confirmed by NOE
experiments (data not shown). The stereochemical control
of these reactions probably arises from a kinetic preference
for attack of hydride upon the R-face of the C-1 carbocation,
producing the â-aryl C-nucleoside. Loss of some stereocon-
trol indicates that the cyclic disiloxane group is affecting
the conformational preferences of the carbocation intermedi-
ate.
represents an efficient method for the syntheses of 1,2-
dideoxy-1-â-aryl-D-ribofuranoses. The reaction sequence
described herein utilizes a stable intermediate that can be
converted in moderate yields and excellent stereoselectivities
into the desired aryl C-nucleosides. The use of other
organolithium reagents in this protocol is currently being
explored.
Acknowledgment. The authors thank the University of
Alabama for support of this project through a RAC grant.
Conversion of disiloxanes 3a-g into the unblocked
C-nucleosides 1a-g can be readily accomplished using
tetrabutylammonium fluoride in THF. All compounds pro-
duced spectroscopic data consistent with the assigned
structures or that agreed with data previously reported by
others1,4a (see Supporting Information). Overall, this route
Supporting Information Available: Preparative proce-
dures and spectroscopic data for 2a, 3a-g, and 1a-g. This
material is available free of charge via the Internet at
http://pubs.acs.org.
OL990813W
Org. Lett., Vol. 1, No. 8, 1999
1175