Spirocyclic restriction of nucleosides. An analysis of protecting group feasibility while accessing prototype anti-1-oxaspiro[4.4]nonanyl mimics

Article abstract of DOI:10.1021/ol010212g

matrix presented The first spirocyclic nucleoside featuring a β-hydroxyl (anti) at C5′ has yielded to synthesis. While the OMOM functionality proved to be sensitive to the conditions necessary to incorporate heterocyclic bases, PMB protection of the carbi

Full text of DOI:10.1021/ol010212g

ORGANIC
LETTERS
2001
Vol. 3, No. 25
4043-4045
Spirocyclic Restriction of Nucleosides.
An Analysis of Protecting Group
Feasibility while Accessing Prototype
anti-1-Oxaspiro[4.4]nonanyl Mimics
Leo A. Paquette,* Dafydd R. Owen, Richard Todd Bibart, and
Christopher K. Seekamp
EVans Chemical Laboratories, The Ohio State UniVersity, Columbus, Ohio 43210
paquette.1@osu.edu
Received September 18, 2001
ABSTRACT
The first spirocyclic nucleoside featuring a â-hydroxyl (anti) at C5′ has yielded to synthesis. While the OMOM functionality proved to be
sensitive to the conditions necessary to incorporate heterocyclic bases, PMB protection of the carbinol was readily accommodated. The
remarkably similar minimum-energy conformations of the title compounds relative to natural thymidine as deduced by Amber calculations in
the gas phase are noted.
The torsion angles around the bond of the sugar-phosphate
DNA backbone are of decisive importance for the secondary
structure of DNA as well as for base recognition.1 Effective
inhibitors of protein expression in vivo have also to resist
the action of DNA-degrading enzymes and bind to their
target mRNA sequences with high affinity and in a sequence-
specific manner.2 Since three sequential enzymatic steps are
required to convert the nucleoside to its 5′-triphosphate, the
complexity of the anabolic process is also an issue. Opti-
mally, identification of the conformational preferences
exhibited by the nucleoside at each step is warranted.
Although idealized conformations have been identified,3 the
inherent flexibility of the furanose ring, which normally
equilibrates rapidly in solution between two extreme forms
of ring pucker,4 often constitutes a major obstacle to
identification of the idealized conformation for a particular
interaction.
In the preceding paper,5 we introduced the concept of
spirocyclic restriction. For the 1-oxaspiro[4.4]nonane series
represented by B and C, fully minimized Amber calculations
in the gas phase have revealed a striking similarity to the
low-energy conformation adopted by natural thymidine (A,
Figure 1). The overlay of B on A is notably remarkable
(RMS ) 0.007). For C, the departure from direct super-
imposition is greater (RMS ) 0.058) but still very acceptable.
In both spirocyclic analogues, the hydroxyl substituent on
the cyclopentane ring is expectedly disposed pseudoequa-
torially. Consequently, the structural change evidenced in
proceeding from B (γ ) +sc) to C (γ ) ap) holds
(1) Eschenmoser, A.; Dobler, M. HelV. Chim. Acta 1992, 75, 218.
(2) (a) Uhlmann, E.; Peyman, A. Chem. ReV. 1990, 90, 543. (b) Milligan,
J. F.; Matteucci, M. D.; Martin, J. C. J. Med. Chem. 1993, 36, 1923.
(3) Van Roey, P.; Taylor, E. W.; Chu, C. K.; Schinazi, R. F. Ann. N. Y.
Acad. Sci. 1990, 616, 29.
(4) Altona, C.; Sundaralingam, M. J. Am. Chem. Soc. 1972, 94, 8205.
10.1021/ol010212g CCC: $20.00 © 2001 American Chemical Society
Published on Web 11/10/2001

Scheme 1
Figure 1. The minimum-energy conformations of thymidine (A),
B, and C.
considerable interest. The torsion angles for the homo dimers
of A-C are compiled in Table 1. Herein, we report the first
dride and pyridine. Failure to execute this protocol leads
instead to tight complexation of the aluminum ion to the
triad of hydroxyl groups. For formation of the spirocyclic
nucleoside analogues, we chose to involve the Lewis acid-
induced coupling of 3 to persilylated thymine7 as in the syn
series.5 The three chromatographically separated products
generated in this reaction were produced in relative yields
that varied widely as a function of the quality of the
trimethylsilyl triflate. The greater the level of adventitious
triflic acid that develops upon prolonged storage of this
reagent (through hydrolysis and degradation), the more
elevated are the proportions of 5 and especially 6. The finding
that the structurally unusual 6 could be made predominant
provided support for the conclusion that oxonium ion D can
indeed be generated transitorily. Irreversible capture of the
activated nucleoside base subsequently materializes.
Table 1. Relevant Torsional Angles for [Thymidine]2, B2, and
C2 (Values in Degrees)
angle
A2
B2
C2
R
â
γ
δ
ꢀ
-63.1
178.4
58.2
126.2
175.3
-85.9
-57.4
116.6
45.6
144.4
-171.3
-37.4
124.7
-89.7
152.7
76.3
-97.0
-88.0
ú
successful access to nucleosides of type B and thereby extend
the range of candidates amenable to further scrutiny. Unlike
the syn series where the MOM protecting group was well
tolerated,5 this substituent proved to be more problematic
when dealing with the somewhat more elevated steric
congestion resident in the anti diastereomers.
Our primary interest was the possible inhibitory steric
effect induced by the protected 5′-hydroxyl on proper
formation of the nucleosidic bond. Initially, attention was
paid to 1, a spirobutenolide building block available in
racemic and enantiopure forms.6 Although the efficiency of
the dihydroxylation of 1 proved to be somewhat variable,
this transformation provided a reliable means for establishing
the C2′-C3′ configurational pattern defined in 2 (Scheme
1). The success of the ensuing reductive acetylation with
generation of 3 rests significantly on a direct quench of the
reaction of the Dibal-H reaction mixture with acetic anhy-
The mode of reaction observed for 3 is not limited to this
specific spirocyclic pseudosugar but appears to be general
for the â-oriented MOM series. Another representative
example, displayed in Scheme 2, consists of the reaction of
lactone 7 with bis(trimethylsilyl)adenine (8)8 in the presence
of trimethylsilyl triflate under otherwise comparable condi-
tions. In this manner, ready access was gained to 9. To our
(5) Paquette, L. A.; Bibart, R. T.; Seekamp, C. K.; Kahane, A. L. Org.
Lett. 2001, 3, 4039.
(6) Paquette, L. A.; Owen, D. R.; Bibart, R. T.; Seekamp, C. K.; Kahane,
A. L.; Lanter, J. C.; Corral, M. A. J. Org. Chem. 2001, 66, 2828.
(7) Vorbru¨ggen, H.; Bennua, B. Chem. Ber. 1981, 114, 1279.
(8) Nishimura, T.; Iwai, I. Chem. Pharm. Bull. 1964, 12, 352.
4044
Org. Lett., Vol. 3, No. 25, 2001

Scheme 2
Scheme 3
knowledge, 6 and 9 represent prototypes of a unique series
of nucleoside mimics that have not been accorded prior
attention.9
To skirt this particular reactivity profile, the replacement
of OMOM by an OPMB group was probed. Triacetate 10
was prepared in the predescribed manner from the levoro-
tatory (>99% enantiomeric excess) butenolide6 (Scheme 3).
The first nucleosidation to be examined involved excess bis-
(trimethylsilyl)cytosine8 in the presence of 1 equiv of
trimethylsilyl triflate in CH2Cl2 as the solvent. The formation
of 11 appeared to fail until recourse was made instead to
use 1,2-dichloroethane as the reaction medium, in line with
the recommendation of others.10 Under these circumstances,
11 could be readily isolated in an unoptimized 60% yield.
No such solvent sensitivity was noted with persilylated uracil,
which gave 12 as the sole nucleosidic product in 54% yield.
In neither reaction was evidence obtained regarding possible
complications arising from the presence of the OPMB
substituent. While the removal of this protecting group by
oxidation with ceric ammonium nitrate failed, excellent
selectivity was realized through the use of DDQ. After mild
saponification, the fully deprotected triol 13, mp 209 °C,
became available as the first example of a 5′â spirocyclic
nucleoside.
In summary, we have defined and prepared representatives
of a new type of conformationally restricted nucleoside. Since
DNA is, like RNA, involved in a wide array of biological
functions, deoxy congeners of the spirocyclic type need also
to be targeted for deliberate incorporation in a site-specific
manner into nucleotide strands. This line of research is being
actively pursued.
Acknowledgment. D.R.O thanks GlaxoWellcome for a
postdoctoral fellowship award (1998-1999) and Prof. Robert
Coleman and Dr. Jason McCary for assistance with the
Amber calculations.
Supporting Information Available: Representative ex-
perimental procedures and select spectral characterization for
the compounds reported herein. This material is available
free of charge via the Internet at http://pubs.acs.org.
(9) For a related observation involving an acyclic methylene acetal,
consult: Chamberlain, S. D.; Biron, K. K.; Dornsife, R. E.; Averett, D. R.;
Beauchamp, L.; Koszalka, G. W. J. Med. Chem. 1994, 37, 1371.
(10) For example: Steffens, R.; Leumann, C. HelV. Chim. Acta 1997,
80, 2426.
OL010212G
Org. Lett., Vol. 3, No. 25, 2001
4045