Synthesis of α-C-glycosides via tandem Tebbe methylenation and Claisen rearrangement

Article abstract of DOI:10.1016/S0040-4039(03)00710-X

A variety of α-C-glycosides may be accessed in an entirely stereoselective fashion from 3-OH glycal esters, by way of the tandem reaction sequence of Tebbe methylenation and Claisen rearrangement. In contrast with previous studies in the corresponding β-series, careful control of conditions for Claisen rearrangement is required in order to avoid loss of integrity of anomeric stereochemistry; thermal rearrangements are best carried out in xylene in a sealed tube.

Full text of DOI:10.1016/S0040-4039(03)00710-X

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 3631–3635
Synthesis of a-C-glycosides via tandem Tebbe methylenation and
Claisen rearrangement
H. Yasmin Godage and Antony J. Fairbanks*
Dyson Perrins Laboratory, Oxford University, South Parks Road, Oxford OX1 3QY, UK
Received 7 February 2003; revised 28 February 2003; accepted 14 March 2003
Abstract—A variety of a-C-glycosides may be accessed in an entirely stereoselective fashion from 3-OH glycal esters, by way of
the tandem reaction sequence of Tebbe methylenation and Claisen rearrangement. In contrast with previous studies in the
corresponding b-series, careful control of conditions for Claisen rearrangement is required in order to avoid loss of integrity of
anomeric stereochemistry; thermal rearrangements are best carried out in xylene in a sealed tube. © 2003 Elsevier Science Ltd.
All rights reserved.
The synthesis of C-glycosides1 remains a significant
occupation of the synthetic community.2 This may be
as part of natural product total synthesis, or in order to
furnish glycomimetics as tools for glycobiology or as
potential therapeutic agents. In particular, stereocon-
trolled access to a wide variety of C-glycosides would
provide materials for biological screening programs,
which may shed further light on the proposed ability of
C-glycosides to act as non-hydrolysable mimics of their
natural O-linked counterparts.3
ture procedures.7 In addition the corresponding
4,6-O-silyl protected compound 3 was itself accessed
from allal 48 by regioselective silylation by treatment
with di-tert-butylsilyl ditriflate in DMF at low tempera-
ture (77% yield, Scheme 1).
Treatment of glycal 1 with benzoyl chloride and cata-
lytic DMAP in pyridine furnished the benzoate ester 5a,
acetylation with acetic anhydride in pyridine yielded the
acetate 5b, whilst treatment with either palmitic acid or
Boc protected 4-amino butyric acid, together with DCC
and catalytic DMAP, correspondingly gave the palmitic
5c and amino butyric esters 5d, respectively. Similarly
glycal 2 was converted into its benzoate 6 by treatment
with benzoyl chloride, whilst silyl protected glycal 3 was
converted into palmitic ester 7. Methylenation of all
esters proceeded smoothly by treatment with an excess
As part of our ongoing studies into the development of
an approach designed at allowing access to a wide
variety of C-glycosides in a parallel synthetic manner,
we recently reported4 the use of glucal derived carbohy-
drate esters to allow access to a variety of b-C-gly-
cosides in high yield and with complete control of
anomeric stereochemistry. Herein we report studies
allowing synthetic access to the corresponding a-C-gly-
cosides, again by use of a tandem reaction sequence
involving Tebbe methylenation5 of glycal esters fol-
lowed by Claisen rearrangement.6
The synthesis of the corresponding a-C-glycosides nec-
essarily entails the use of glycal esters derived from
allose, which is epimeric at the 3-position to glucose. To
this end we undertook the synthesis of the known
4,6-O-benzylidene protected allal 1 and the correspond-
ing C-2 methyl substituted derivative 2, following litera-
Keywords: C-glycosides; carbohydrates; glycals; Tebbe reagent;
Claisen rearrangement.
* Corresponding author. Tel.: +441865275647; fax: +441865275674;
e-mail: antony.fairbanks@chem.ox.ac.uk
t
Scheme 1. Reagents and conditions: (i) Bu2Si(OTf)2, DMF,
−40°C to rt, 77%.
0040-4039/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0040-4039(03)00710-X

3632
H. Y. Godage, A. J. Fairbanks / Tetrahedron Letters 44 (2003) 3631–3635
Scheme 2. Reagents and conditions: (i) PhCOCl, DMAP, pyridine, 0°C; (ii) Ac2O, pyridine, rt; (iii) RCO2H, DCC, DMAP,
CH2Cl2, rt; (iv) Tebbe, THF:pyridine, 4:1, −40°C.
of Tebbe reagent (ꢀ2 equivalents) at −40°C in a mix-
ture of THF and pyridine (ratio ꢀ4:1), to yield the
corresponding enol ethers 8–10 (Scheme 2).
tial advantage of the tandem Tebbe/Claisen approach
to C-glycoside synthesis over alternative intermolecular
approaches resided solely in the ability to furnish
totally anomerically pure products, the search for reac-
tion conditions that could yield pure a-C-glycosides
became somewhat of a prerogative.
With a selection of vinyl ethers in hand attention
turned to the subsequent Claisen rearrangement. Ther-
mal rearrangement of enol ether 8a, was attempted
following conditions that had successfully yielded the
corresponding b-C-glycosides.4 However it was found
that heating 8a to 180°C in benzonitrile as solvent
produced a mixture of both a- and b-C-glycoside prod-
ucts 11a and 12, albeit in a favourable ratio of a:b 5:1
(Scheme 3). In addition small amounts of the open
chain diene 139 were also observed, indicating the prob-
able mechanism by which the a- and b-C-glycosides
interconverted.10 Although in the previously studied
b-series the minor amounts of epimerisation that had
occasionally been observed during thermal rearrange-
ment could be suppressed by changing the solvent to
tri-n-butylamine, in the a-series this proved not to be
the case and once again mixtures of anomers were
formed.
Lewis acid catalysed reactions were investigated as an
alternative to thermal rearrangement. A selection of
Lewis acids, including NaBF4, AlCl3, BF3·OEt2,
Yb(OTf)3, and TiCl4 was screened under a variety of
reaction conditions, but unfortunately in all cases mix-
tures of epimeric products were formed, together with
variable amounts of open chain diene 13. Although in
Although these observations were in contrast to litera-
ture accounts of a similar Claisen rearrangement which
had not reported competitive formation of the b-
anomer,6e related work had indeed indicated the ease
by which the a-C-glycosides products may be
epimerised to their thermodynamically more favoured
b-counterparts.11 It was therefore clear that if this
competitive anomerisation process were to be sup-
pressed then careful choice and control of reaction
conditions would be required. However since the poten-
Scheme 3. Reagents and conditions: (i) Bu3N, 180°C; (ii)
PhCN, 180°C; (iii) various Lewis acids at low temperature.

H. Y. Godage, A. J. Fairbanks / Tetrahedron Letters 44 (2003) 3631–3635
3633
Figure 1. Crystal structure of C-glycoside 11a showing crystallographic numbering scheme [thermal ellipsoid plot (ORTEP-34) at
40% probability].
some cases the ratio of products was favourable
towards the desired a-anomer (e.g. a:b ratio of 16:1, for
BF3·OEt2), the yields for these reactions were at best
modest (ꢀ60%) and since the goal was complete con-
trol of anomeric stereochemistry alternative reaction
conditions were continually pursued.
sealed tube. Rather surprisingly when the thermal rear-
rangement of 11a was performed in d6-benzene in a
sealed tube the reaction proceeded smoothly and no
formation of the undesired i-anomer was observed. This
initially pointed the way to the use of benzene as the
solvent for the rearrangement reaction. Indeed a series
of thermal rearrangements in benzene did produce pure
a-C-glycosides. However mindful of the undesirable
properties of benzene, a selection of alternative solvents
were screened at a variety of temperatures. Thermal
rearrangement in xylene at 195°C proved to be opti-
mum, and gratifyingly enol ethers 8a, 8b and 9 all
In the face of the persistent anomerisation problem it
was thought prudent to monitor the rate of formation
of the undesired b-anomer relative to the Claisen rear-
rangement. To this end a series of NMR reactions were
performed in deuterated benzene as the solvent in a
Scheme 4. Reagents and conditions: (i) xylene, sealed tube, 195°C; (ii) xylene, sealed tube, 195°C, ratio 11c:16, 1.3:1, yield 85%;
(iii) d6 benzene, 195°C, sealed tube, ratio 11c:16, 1:1, yield 84%; (iv) xylene, sealed tube, 195°C, 15 only, yield 85%; (v) d6 benzene,
195°C, sealed tube, ratio 15:17, 1.4:1, yield 94%.

3634
H. Y. Godage, A. J. Fairbanks / Tetrahedron Letters 44 (2003) 3631–3635
rearranged smoothly in excellent yield and most impor-
tantly, entirely stereoselectively, to yield only the a-C-
glycoside products 11a, 11b, and 14 respectively
(Scheme 4). The anomeric stereochemistry of C-gly-
coside 11a was confirmed by X-ray crystallography
(Fig. 1),12 whilst the anomeric configuration of the
other a-C-glycosides was confirmed by NOE difference
experiments.13
Tetrahedron Organic Chemistry Series Volume 13, Perga-
mon Press: Oxford, 1995.
2. For some recent references, see: (a) Chiara, J. L.; Sesmilo,
E. Angew. Chem., Int. Ed. Engl. 2002, 41, 3242–3246; (b)
Paterson, D. E.; Griffin, F. K.; Alcaraz, M.-L.; Taylor,
R. J. K. Eur. J. Org. Chem. 2002, 1323–1336; (c) Abe, H.;
Shuto, S.; Matsuda, A. J. Am. Chem. Soc. 2001, 123,
11870–11882; (d) Singh, G.; Vankayalapati, H. Tetra-
hedron: Asymmetry 2001, 12, 1727–1735.
However Boc protected amine 8d reacted only very
slowly under these conditions, and no appreciable
amount of product was observed-the starting material
being recovered in this case. Moreover during rear-
rangement of the two palmitic esters 8c and 10 to
produce the desired a-C-glycosides 11c and 15 the
formation of two side products was occasionally
observed. These side products were identified as the
a-C-glycosides 16 and 17,14 and are presumably formed
via partial isomerisation of the glycal enol ethers 8c and
10 to the thermodynamically preferred more substituted
tautomers before rearrangement. Frustratingly the rela-
tive amounts of these products formed appeared to be
quite variable. For example in one instance the use of
xylene as solvent for rearrangement of 10 completely
suppressed the formation of 17, whilst during a similar
NMR experiment with d6 benzene as solvent resulted in
the formation of 17 in an almost equal amount to that
of the desired product 15. However the formation of 16
from 8c was observed even in xylene as solvent. Com-
plete suppression of the formation of these products in
this case has not yet proved possible,15 although further
investigations are ongoing.
3. For leading references on conformational comparisons
between O- and C-glycosides, see: (a) Jimenez-Barbero,
J.; Espinosa, J. F.; Asensio, J. L.; Canada, F. J.; Poveda,
A. Adv. Carb. Chem. Biochem. 2001, 56, 235–284; (b)
O’Leary, D. J.; Kishi, Y. J. Org. Chem. 1994, 59, 6629–36
and references contained therein.
4. Godage, H. Y.; Fairbanks, A. J. Tetrahedron Lett. 2000,
41, 7589–7593.
5. Tebbe, F. N.; Parshall, G. W.; Reddy, G. S. J. Am.
Chem. Soc. 1978, 100, 3611–3613.
6. For alternative approaches to C-glycosides by Claisen
rearrangement, see: (a) Ireland, R. E.; Wilcox, C. S.;
Thaisrivongs, S.; Vanier, R. Can. J. Chem. 1979, 57,
1743–1745; (b) Fraser-Reid, B.; Dawe, R. D.; Tulshian,
D. B. Can. J. Chem. 1979, 57, 1746–1749; (c) Ireland, R.
E.; Wuts, P. G. M.; Ernst, B. J. Am. Chem. Soc. 1981,
103, 3205–3207; (d) Ireland, R. E.; Anderson, R. C.;
Badoub, R.; Fitzsimmons, B. J.; McGarvey, G.; Thais-
rivongs, S.; Wilcox, C. S. J. Am. Chem. Soc. 1983, 105,
1988–2006; (e) Tulshian, D. B.; Fraser-Reid, B. J. Org.
Chem. 1984, 49, 518–522; (f) Curran, D. P.; Suh, Y. G.
Carbohydr. Res. 1987, 111, 161–191; (g) Colombo, L.;
Casiraghi, G.; Pittalis, A.; Rassu, G. J. Org. Chem. 1991,
56, 3897–3900; (h) Vidal, T.; Haudrechy, A.; Langlois, Y.
Tetrahedron Lett. 1999, 40, 5677–5680; (i) Wallace, G. A.;
Scott, R. W.; Heathcock, C. H. J. Org. Chem. 2000, 65,
4145–4152.
In summary, the tandem Tebbe methylenation and
thermal Claisen rearrangement has been extended to
allow the synthesis of a variety of a-C-glycosides.16 In
particular to obtain pure a-products careful control of
reaction conditions is required in order to avoid com-
peting formation of the thermodynamically more stable
b-C-glycoside products. This is currently best achieved
by performing the thermal reactions in either xylene or
benzene as solvent in a sealed tube, though in the case
of palmitic enol ethers isomerisation prior to rearrange-
ment is found to be a competing process. Further
investigations into the use of the tandem Tebbe/Claisen
approach for the synthesis of a wide variety of C-gly-
cosides, C-glycosyl amino acids and C-oligosaccharides
as potential glycomimetics are currently in progress and
results will be reported in due course.
7. (a) Sharma, M.; Brown, R. K. Can. J. Chem. 1966, 44,
2825–2835; (b) Sharma, M.; Brown, R. K. Can. J. Chem.
1968, 46, 757–766; (c) for data see: Guthrie, R. D.; Irvine,
R. W. Carbohydr. Res. 1979, 72, 285–288.
8. Allal (see Ref. 7c) was itself synthesised from diacetone
glucose, by a sequence of oxidation and stereoselective
reduction. Full details will be published separately.
9. The EE configuration of diene 13 is suggested by the
coupling constants between H-2 and H-3, and H-4 and
H-5 which were both ꢀ15 Hz (ketone numbered as C-1).
10. Epimerisation probably occurs through a retro-Michael
reaction of the desired a-C-glycoside to produce the open
chain diene, which is then followed by re-closure of the
5-hydroxyl group onto the alternative face of the alkene.
11. Dawe, R. D.; Fraser-Reid, B. J. Org. Chem. 1984, 49,
522–528.
Acknowledgements
12. Crystal data for 11a: C21H20O4; M=336.39; monoclinic;
space group P21; a=4.7401(4), b=10.8660(7), c=
We gratefully acknowledge financial support from the
EPSRC (Project Studentship to H.Y.G.) and the use of
the EPSRC Mass Spectrometry Service (Swansea, UK)
and the Chemical Database Service (CDS) at Dares-
bury, UK.
,
16.1935(14) A; h=90.00, i=92.686, k=90.00°; cell vol-
3
3
,
ume=833.1 A , Z=2; calculated density=1.341 Mg/m ;
R=0.0335; wR=0.0366. Diffraction data were measured
at 150 K using an Enraf–Nonius Kappa CCD diffrac-
tometer (graphite-monochromated MoKa radiation, u=
,
0.71073 A). Intensity data were processed using the
References
DENZO-SMN package. Crystallographic data (excluding
structure factors) has been deposited with the Cambridge
1. Levy, D. E.; Tang, C. The Chemistry of C-Glycosides,

H. Y. Godage, A. J. Fairbanks / Tetrahedron Letters 44 (2003) 3631–3635
3635
Crystallographic Data Centre as supplementary publica-
tion number CCDC 203253. Copies of the data can be
obtained free of charge, on application to CCDC, 12
Union Road, Cambridge CB2 1EZ, UK [fax: +44(0)-
1223-336033 or e-mail: deposit@ccdc.cam.ac.uk].
13. In all cases NOE difference experiments revealed no
enhancements between H-1 and H-5 (original carbohy-
drate numbering), whilst enhancements were observed
between H-5 and the two protons of the anomeric meth-
ylene carbon.
been possible to unequivocally assign their stereochem-
istry.
15. Indeed it has subsequently been discovered that sealed
tube rearrangement of palmitic esters in the b-series
(detailed in Ref. 4) in xylene at 195°C also leads to the
formation of these isomeric products, whereas none were
previously observed during thermal rearrangement at
180°C in either benzonitrile or tributylamine.
16. All new compounds possess spectroscopic data consistent
with their structures, together with satisfactory microana-
lytical and/or high-resolution mass spectral data.
14. Both are formed as single diastereomers but it has not yet