Efficient Route to 2-Deoxy β-O-Aryl-D-Glycosides via Direct Displacement of Glycosyl Iodides

Article abstract of DOI:10.1021/ol035705v

(Equation presented) The conversion of glycals to 2-deoxy glycosyl acetates followed by reaction with trimethylsilyl iodide affords the corresponding glycosyl iodides, which readily undergo substitution with aryl alkoxy anions to provide 2-deoxy-β-O-aryl

Full text of DOI:10.1021/ol035705v

ORGANIC
LETTERS
2003
Vol. 5, No. 22
4219-4222
Efficient Route to 2-Deoxy
â-O-Aryl-D-Glycosides via Direct
Displacement of Glycosyl Iodides
Son N. Lam and Jacquelyn Gervay-Hague*
Department of Chemistry, UniVersity of California, DaVis,
One Shields AVe., DaVis, California 95616
gerVay@chem.ucdaVis.edu
Received September 5, 2003
ABSTRACT
The conversion of glycals to 2-deoxy glycosyl acetates followed by reaction with trimethylsilyl iodide affords the corresponding glycosyl
iodides, which readily undergo substitution with aryl alkoxy anions to provide 2-deoxy-â-O-aryl glycosides. Direct displacement of the anomeric
iodide alleviates the need to introduce temporary C-2 stereodirecting groups that require subsequent removal. The only observable byproducts
from the glycosylations result from elimination of HI giving the starting glycals, which can be recycled through the reaction sequence.
Deoxygenated carbohydrates are major structural components
of numerous natural products possessing important biological
properties.1 For example, 2,6-dideoxy hexopyranoses are
common structural units of olivomycin (1) and chromomycin
(2) antitumor antibiotics (Figure 1). These compounds are
of interest not only due to their biological profiles but also
because of the synthetic challenges they present. With a
methylene at C-2, there is no apparent functionalizable handle
with which to direct glycosylation.
group often requires toxic reagents, which introduce impuri-
ties that are difficult to remove. Moreover, this approach
involves additional steps that can be avoided with direct
addition strategies (vide infra).4
Falck and co-workers have employed triphenylphosphine-
hydrogen bromide (TPHB) as a soft catalyst to promote direct
addition of various acceptors to glycals.5 These methodolo-
gies successfully circumvent Ferrier rearrangement and
Two general strategies for entry into 2-deoxy glycosides
have been pursued.2 Most commonly, a directing group is
temporarily introduced at C-2, the acceptor is then incorpo-
rated, and in a third step the C-2 substituent is exchanged
with hydrogen (Figure 2).3 Removal of the C-2 directing
(1) (a) Kirsching, A.; Bechthold, A. F.-W.; Rohr, J. Top. Curr. Chem.
1997, 188, 1. (b)Weymouth-Wilson, A. C. Nat. Prod. Rep. 1997, 14, 99.
(2) (a) Veyrieres, A. Carbohydr. Chem. Biol. 2000, 1, 367. (b) Franck,
R.; Marzabadi, C. Tetrahedron 2000, 56, 8385.
(3) (a) Takiura, K.; Honda, S. Carbohydr. Res. 1972, 23, 369. (b)
Monneret, C.; Choay, P. Carbohydr. Res. 1981, 96, 299. (c) Giese, B.;
Gilges, S.; Groninger, K. S.; Lamberth, G.; Witzel, T. Liebigs Ann. Chem.
1988, 615. (d) Geise, B.; Kopping, B. Tetrahedron Lett. 1989, 30, 681. (e)
Thiem, J.; Klafke, W. Top. Curr. Chem. 1990, 154, 285. (f) Gervay, J.;
Danishefsky, S. J. Org. Chem. 1991, 56, 5448. (g) Roush, W. R.; Briner,
K.; Sebesta, D. P. Synlett. 1993, 264. (h) Roush, W. R.; Bennett, C. E. J.
Am. Chem. Soc. 1999, 121, 3541. (i) Roush, W. R.; Narayan, S. Org. Lett.
1999, 1, 899. (j) Roush, W. R.; Durham, T. B. Org. Lett. 2003, 5, 1871.
Figure 1.
10.1021/ol035705v CCC: $25.00 © 2003 American Chemical Society
Published on Web 10/07/2003

â-iodide or by axial addition to an oxacarbenium intermedi-
ate. Conversely, strong nucleophiles typically yield â-gly-
cosides by SN2-like attack on the thermodynamically more
stable R-glycosyl iodide. We were encouraged by the report
that glycosyl bromides yield â-glycosides when activated
with insoluble silver catalysts,9 and we anticipated that
glycosyl iodides would offer clear advantages in direct
displacement reactions, including the ability to do reactions
under anionic conditions, with increased stereoselectivities,
and decreased reaction times.10
Glycosyl iodides are efficiently prepared in situ from the
corresponding anomeric acetate upon reaction with trimeth-
ylsilyl iodide (TMSI). We therefore set out to prepare
2-deoxy and 2,6-dideoxy glycosyl acetates by the direct
addition of acetic acid to glycals (Scheme 1). In our initial
Figure 2.
directly introduce hydrogen at C-2, but R-glycosides often
predominate due to the kinetic anomeric effect.6 More
recently, studies by McDonald and co-workers suggest that
substituent effects may moderate the stereochemical outcome
in these reactions.7
Scheme 1
As a part of our ongoing investigations probing the utility
of glycosyl iodides,8 it occurred to us that 2-deoxy-â-
glycosides may be readily accessible via direct displacement
of R-glycosyl iodides, which in turn would be derived from
glycals (Figure 3). Our previous studies have shown that
Figure 3.
investigations, treatment of 3 with stoichiometric amounts
of anhydrous hydrogen iodide, generated in situ from TMSI
and anhydrous acetic acid, quickly provided 4 as well as
hemiacetal 5. When equimolar hydrogen bromide, produced
from acetyl bromide and anhydrous acetic acid, was used, a
mixture of 4 and 5 was also obtained in comparable time to
the hydrogen iodide assisted approach. Both conversions
reaction conditions promoting in situ anomerization provide
R-glycosides either by nucleophilic displacement of the
(4) (a) Bielawska, H.; Michalska, M. J. Carbohydr. Chem. 1986, 5 (3),
445. (b) Roush, W. R.; Lin, X.-F. J. Org. Chem. 1991, 56, 5740. (c)
Andrews, F. L.; Larsen, D. S. Tetrahedron Lett. 1994, 35, 8693. (d)
Andrews, F. L.; Larsen, D. S., Larsen, L. Aust. J. Chem. 2000, 53, 15.
(5) Bollit, V.; Mioskowski, C.; Lee, S.-G.; Flack, J. R. J. Org. Chem.
1990, 55, 5812.
(8) (a) Gervay, J.; Nguyen, T. N.; Hadd, M. J. Carbohydr. Res. 1997,
300, 119. (b) Hadd, M. J.; Gervay, J. Carbohydr. Res. 1999, 320, 61. (c)
Gervay, J. Org. Synth.: Theory Appl. 1998, 4, 121. (d) Bhat, A. S.; Gervay-
Hague, J. Org. Lett. 2001, 3, 2081.
(6) Kaila, N.; Blumenstein, M.; Bielawska, H.; Franck, R. W. J. Org.
Chem. 1992, 57, 4576.
(7) McDonald, F. E.; Subba Reddy, K. Angew. Chem., Int. Ed. 2001,
40, 3653.
(9) Binkley, R. W.; Koholic, D. J. J. Org. Chem. 1989, 54, 3577.
(10) Gervay, J.; Hadd, M. J. J. Org. Chem. 1997, 62, 6961.
4220
Org. Lett., Vol. 5, No. 22, 2003

typically gave high overall mass recovery, and although 5
predominated in these reactions, acetylation quickly furnished
4. These results are in accord with the belief that softer acids
suppress the Ferrier I, as allylic rearrangement products were
not detected.5
Scheme 3
To simplify our strategy, we chose to employ com-
mercially available 30 wt % HBr/HOAc and added acetic
anhydride to the reaction mixture to suppress the formation
of 5. Using substoichiometric amounts of HBr in acetic acid
(25% molar to the glycal) in combination with anhydrous
acetic anhydride, we were able to generate primarily 4 in
high yields, again without rearranged products. Treatment
of 3 with catalytic TMSI in HOAc and Ac2O also afforded
the 2-deoxy-tetra-O-acetate (4) in comparable time. However,
after aqueous workup and chromatography, 4 was recovered
in only 45% yield, as C-1 hydrolysis had also occurred. Using
0.1 molar equiv of HBr/HOAc with respect to the glycal
also provided 4, but longer times, 4 days, were required for
completion. Since HBr/HOAc is cheaper than TMSI and
there were no obvious advantages to using HI generated in
situ, we opted for the more cost-efficient route.
The reactions were performed on multigram scale, and the
workup was facilitated by the addition of sodium acetate
(NaOAc) to neutralize HBr, resulting in NaBr precipitation.
Filtration of the salts, rotoevaporation to dryness, and
recrystallization from ethyl acetate-hexanes supplied the
tetra-O-acetate 4 (>90%) in short order. The general
applicability of this method was further illustrated by the
reactions of 6 and 8, providing 2-deoxygalactosyl acetate 7
and 2,6-dideoxyglucosyl acetate 9, respectively.
With a simple route affording large quantities of building
blocks in hand, formation of 2-deoxy glycosyl iodides by
the method of Thiem and Meyer was explored.11 As we
anticipated, treatment of 4 with TMSI at 0 °C for 15 min in
dichloromethane afforded 1-iodo-2-deoxy-3,4,6-tri-O-acetyl-
R-D-arabino-pyranoside (10). Quantitative conversion to the
R-iodide was evidenced by 1H NMR data showing a dramatic
downfield shift of the anomeric proton (δ 6.95 ppm) when
compared with the anomeric acetate (δ 6.25 ppm).12 Glycosyl
iodides 11 and 12 were synthesized in a similar fashion
(Scheme 2).
o-cresolate or naphthoate anions to the 2-deoxyglucosyl 10
and 2-deoxygalactosyl 11 iodides was complete within 15
min, producing the â-O-aryl-glycosides 13, 14, 16, and 17
as the only detectable glycosylation products.
The anomeric configuration was assigned on the basis of
C1-H1 coupling constants obtained13 from a modified
HSQC experiment.14 By lowering the 1H-decoupling power,
the 1J and 2J resonances were detected.15 A one-dimensional
1
2
slice from the F2 axis revealed J and J couplings of the
anomeric carbon and proton. For example, at 500 MHz, the
1J C1-H1 coupling of 16 was calculated to be 160.3 Hz
2
1
and the J large H1-H2ax was 9.8 Hz. The J C1-H1
1
coupling was identical to those calculated from data of H-
coupled full NOE 13C-spectra. However, since the HSQC is
a proton-based experiment, the spectroscopic data were
acquired more quickly and with less material than is required
for 13C experiments.
It is notable that comparable â-selectivity was obtained
for both the glucosyl and galactosyl iodides, as stereochem-
ical control is often diminished when employing 2-deoxy-
galactosyl donors.3j Nucleophilic displacement of 2,6-
dideoxyglucosyl iodide 12 also provided the â-O-aryl-
glycosides 15 and 18 as the only detected glycosylation
products; however, the reactions were not as efficient when
compared to donors 10 and 11. Under the basic conditions
of addition, 1,2-elimination became the alternate pathway,
resulting in regeneration of the starting glycal 8. The
propensity of 12 to undergo elimination is presumably due
to increased electron density relative to 10 and 11, which
have electron-withdrawing groups at C-6. We have previ-
ously reported that stereoelectronic effects attenuate elimina-
tion.16
Scheme 2
(11) Thiem, J.; Meyer, B. Chem. Ber. 1980, 113, 3075.
(12) Experiments were performed in deuteriochloroform at 298 K.
(13) Bock, K.; Lundt, I.; Pedersen, C. Tetrahedron Lett. 1973, 14, 1037.
(14) Willker, W.; Leibfritz, D.; Kerssebaum, R.; Bermel, W. J. Magn.
Res. Chem. 1993, 31, 287.
(15) All two-dimensional NMR experiments were performed on a Bruker
DRX 500 or a Bruker DRX 600 MHz instrument housing a Bruker
Cryoprobe.
(16) (a) Lam, S. N.; Gervay-Hague, J. Org. Lett. 2002, 4 (12), 2039. (b)
Lam, S. N.; Gervay-Hague, J. Carbohydr. Res. 2002, 337, 1953.
The glycosyl iodides were subsequently reacted with
arylalkoxy anions under direct displacement conditions
(KHMDS, 18-crown-6). As shown in Scheme 3, addition of
Org. Lett., Vol. 5, No. 22, 2003
4221

In conclusion, our studies demonstrate that direct addition
of acetic acid to glycals results in high yields of 2-deoxy
glycosyl acetates, which can be readily converted to the
corresponding R-glycosyl iodides in quantitative yield.
Substitution of the iodides with arylalkoxy anions, in what
is presumably an SN2-like displacement, results in highly
stereoselective reactions to afford 1-O-aryl-2-deoxy-â-D-
glycopyranosides. Glycals are the only observable byprod-
ucts, and they can be easily recycled to the corresponding
glycosyl iodide via the 2-deoxy-glycosyl acetate precursors.
The ability to readily recycle the glycal byproducts is an
important advantage given that many natural products contain
rare 2-deoxy-L-sugars.17
Acknowledgment. Support of this work was provided
by NSF CHE-0196482, NSF CRIF program (CHE-9808183),
NSF OSTI 97-24412, and NIH RR11973. We also acknowl-
edge assistance from Dr. Jeff DeRopp and Dr. Janos Csanadi
in implementing HSQC experiments.
Supporting Information Available: Experimental pro-
cedures and complete characterizations for compounds 4 and
7-18. This material is available free of charge via the
Internet at http://pubs.acs.org.
(17) (a) Roush, W. R.; Lin, X.-F. J. Am. Chem. Soc. 1995, 117, 2236.
(b) Durham, T. B.; Roush, W. R. Org. Lett. 2003, 5 (11), 1875.
OL035705V
4222
Org. Lett., Vol. 5, No. 22, 2003