Synthesis of novel C-nucleosides with potential applications in combinatorial and parallel synthesis

Article abstract of DOI:10.1016/S0040-4039(99)02078-X

Novel di-substituted pyrimidinyl C-nucleosides have been synthesised using a flexible synthetic methodology. This methodology conveniently allows the synthesis of alpha and beta nucleosides, both of which are of biological interest. Reactive acetylenic ke

Full text of DOI:10.1016/S0040-4039(99)02078-X

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 575–578
Synthesis of novel C-nucleosides with potential applications in
combinatorial and parallel synthesis
Robert M. Adlington, Jack E. Baldwin,∗ Gareth J. Pritchard and Keith C. Spencer
The Dyson Perrins Laboratory, South Parks Road, Oxford, OX1 3QY, UK
Received 12 October 1999; accepted 2 November 1999
Abstract
Novel di-substituted pyrimidinyl C-nucleosides have been synthesised using a flexible synthetic methodology.
This methodology conveniently allows the synthesis of alpha and beta nucleosides, both of which are of biological
interest. Reactive acetylenic ketones, derived from 2-deoxyribose, can be reacted with substituted amidines to give
new families of C-nucleoside natural product analogues. © 2000 Elsevier Science Ltd. All rights reserved.
We recently reported the use of an acetylenic ketone as a reactive motif for the synthesis of heterocyclic
substituted amino acids.1,2 We now wish to report extension of this methodology to the carbohydrate
area. C-Nucleosides, such as showdomycin,3 show a broad spectrum of biological activity and so
have stimulated considerable interest as potential anti-tumour, anti-bacterial and anti-cancer agents.4
Although most biologically active nucleosides are of the β-configuration,4 there is growing interest in
those of the α-configuration, which show biological activity.5 Therefore, we set out to develop a flexible
synthetic methodology which would allow the synthesis of pyrimidinyl C-nucleosides of both anomeric
configurations. It would be preferable if we could install this feature at a late synthetic stage.
Naturally occurring C-nucleosides are based around either a ribose or 2-deoxyribose core. As
C-nucleosides of 2-deoxyribose are reportedly not as well known in the literature as traditional
2-deoxyribonucleosides we chose the synthetic challenge posed by preparing 2-deoxyribose C-
nucleosides.6 It can be envisaged that reaction of a carbohydrate-substituted acetylenic ketone with
substituted amidines would give a family of novel di-substituted pyrimidinyl C-nucleosides (Fig. 1).7
Choice of protecting group is important since it must permit both the formation of the acetylenic
ketone 2, and be easily removed without destroying the final C-nucleoside targets. Deprotection must
also take into account the benzylic nature of the C1-oxygen bond in the tetrahydrofuran ring. After other
alternatives were eliminated, benzyl protection and BCl3 deprotection proved optimal.
3,5-Di-O-benzyl-2-deoxy-α-D-ribofuranosyl cyanide 5, was synthesised from protected 2-
deoxyribose8,9 using literature precedent (Scheme 1).10 This reaction gave an 81% yield of the
∗
Corresponding author. E-mail: jack.baldwin@chem.ox.ac.uk (J. E. Baldwin)
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(99)02078-X
tetl 16006

576
Fig. 1.
nitrile with predominantly the α stereochemistry (2:1 α:β ratio) at the anomeric centre.11 Only the α
anomer was used in subsequent synthetic steps. Hydrolysis of the nitrile 5, under acidic conditions, gave
the acid 6 in quantitative crude yield. Formation of the Weinreb amide 7 was similarly successful, so
giving the intermediate for the crucial ketone formation step in gram quantities.8,12
Scheme 1. Reagent and conditions: (i) MeOH, AcCl, rt, 7 min then Ag2CO3; (ii) BnBr, NaH, Bu4NI, DMF, −30oC to rt, 24
h; (iii) TMSOTf, TMSCN, DCM, 24 h, rt, then chromatography [silica gel, (5:1) petroleum ether:Et2O]; (iv) HCl, 1,4-dioxan,
100°C, 4 h; (v) α,α-dichloromethylmethyl ether, DCM, reflux, 5 h, then Me(OMe)NH·HCl, NEt3, DCM, rt, 24 h
The Weinreb amide 7 was reacted with a variety of lithium acetylides (LiCCR) which, after mild
acidic work up, yielded ketones 8(a–c) in near quantitative yield (Scheme 2). The acetylenic ketones
were highly sensitive to oxygen and were used without purification in the following cyclisation reaction.
Cyclisation of the ketones, in strictly deoxygenated solvents, with substituted amidines, resulted in the
formation of the desired protected C-nucleosides 9(a–f) in good yields, as the anomeric mixture in a
1:1 α:β ratio. Where R is TMS, 8a, the silicon group was hydrolytically cleaved during reaction to give
the parent compounds, 9a and 9b, R=H. All compounds were purified by column chromatography on
silica gel. After considerable investigation an optimised protocol using BCl3 in DCM for up to 24 hours
at room temperature, was found to smoothly deprotect all compounds (except 9a) giving 10(b–f). All
compounds 10(b–f) were isolated and purified by column chromatography on silica gel.
In a separate study the separated α and β anomers of 9(b–f) epimerised under the deprotection
protocol to give product mixtures 10(b–f) in a 1:1 α:β ratio.
Although the fully deprotected compounds 10(b–f) can be isolated as anomeric mixtures and screened
for biological activity, we also wished to isolate gram quantities of separated anomers. Thus, anomeric
mixtures of compounds 10(c–e) were benzoyl protected,13 the anomers separated chromatographically,
then deprotected under basic conditions, to yield the anomerically pure, fully deprotected C-nucleosides
13 and 14(c–e) in 50–57% combined yield from 9(c–e) (Scheme 3).11,14
To conclude, we have developed and extended a methodology for the synthesis of di-substituted
pyrimidines to the carbohydrate field, giving access to a family of previously unknown C-nucleosides.
They were formed in isolated parallel batches in good overall yield with minimal chromatography. Such
compounds are likely to be predisposed to biological activity and we are presently determining that
activity. Novel carbohydrate-substituted acetylenic ketones have been synthesised, as intermediates, by

577
Scheme 2. Reagents and conditions: (i) LiCCR (2 equiv.), THF, −78oC; 3 h, then NH4Cl(aq), −78oC to rt; (ii) Na2CO3, MeCN,
H2O, 80°C, 24 h; (iii) BCl3, (2.1 equiv.), DCM, rt, 2–24 h
Scheme 3. Reagents and conditions: (i) BCl3, DCM, rt, 2–24 h then BzCl, DCM, DMAP, rt, 24 h; (ii) LiOH, THF, H2O, rt
reaction of an activated carboxylic acid with lithiated acetylides, which upon cyclisation with substituted
amidines, gave a family of C-nucleosides, as the anomeric mixture. Lewis acid catalysed deprotection
was found to epimerise either α- or β-benzyl protected C-nucleosides 9(b–f) to give both the α and β C-
nucleosides as the 1:1 anomeric mixture. Potentially, this methodology allows other heterocyclic systems
to be accessed by reaction of α,β unsaturated acetylenic ketones to give other novel C-nucleosides.
Therefore, we have demonstrated that the parallel synthesis of heterocyclic natural products can be
achieved by the reaction of 2-deoxyribose-derived acetylenic ketones and can thus be applied to the
preparation of novel C-nucleosides.
Acknowledgements
We wish to thank the BBSRC for a studentship to KCS.

578
References
1. Adlington, R. M.; Baldwin, J. E.; Pritchard, G. J.; Catterick, D. J. Chem. Soc., Perkin 1 1999, 855.
2. Adlington, R. M.; Baldwin, J. E.; Pritchard, G. J.; Catterick, D. J. Chem. Soc., Chem. Commun. 1997, 1757
3. (a) Barton, D. H. R.; Ramesh, M. J. Am. Chem. Soc. 1990, 112, 891. (b) Kang, S. O.; Lee, S. B. J. Chem. Soc., Chem.
Commun. 1995, 1017.
4. For reviews, see: Perigaud, C.; Gosselin, G.; Imbach, J.-L. Nucleosides Nucleotides 1992, 11, 903.
5. Bernardo, S. D.; Weigele, M. J. Org. Chem. 1976, 41, 287.
6. Chaudhuri, N. C.; Ren, R. X.-F.; Kool, E. T. Synlett 1997, 341
7. Fox has described related methodology as a route to pyrimidin-4-ones, see: Tam, S. Y.-K.; Heras, F. G. D. L.; Klein, R. S.;
Fox, J. J. Tetrahedron Lett. 1975, 38, 3271.
8. Rolland, V.; Kotera, M.; Lhomme, J. Synth Commun. 1997, 27, 3505.
9. Brimacombe, J. S.; Fletcher, H. G. Methods Carbohydr. Chem. 1972, 6, 373
10. Jung, M. E.; Trifunovich, I.; Gardiner, J.; Clevenger, G. L. J. Chem. Soc., Chem. Commun. 1990, 84.
11. Confirmed by NOE experiments on both epimers.The anomeric configuration of all anomers were confirmed by NOE
experiments based on the known configuration of substituents at C3 and C4.
12. Cupps, T. L.; Boutin, R. H.; Rapoport, H. J. Org. Chem. 1985, 50, 3972.
13. A more direct route to separate α and β anomers using only benzoyl-protecting groups was unsuccessful due to the failure to
form an acetylenic ketone from the benzoyl analogue of 7. Other routes using Lewis acid catalysed addition of trimethylsilyl
acetylenes, and Stille cross-coupling reactions on stannyl acetylenes, all failed.
14. Yields of C-nucleosides, 13 and 14(c–e) were calculated from the starting benzyl protected C-nucleosides, 9(c–e).
Deprotected C-nucleosides 13 and 14(c–e) were purified by column chromatography on silica gel using an appropriate
solvent system.