2982
J . Org. Chem. 1999, 64, 2982-2983
Sch em e 1
Syn th eses of D- a n d L-Ma n n ose, Gu lose, a n d
Ta lose via Dia ster eoselective a n d
En a n tioselective Dih yd r oxyla tion Rea ction s
J oel M. Harris, Mark D. Keranen, and
George A. O’Doherty*
Sch em e 2
Department of Chemistry, University of Minnesota,
Minneapolis, Minnesota 55455
Received March 8, 1999
The de novo enantioselective synthesis of the hexoses
stands as a challenge to asymmetric catalysis.1 Despite
some germinal efforts toward the hexoses, notably by
Masamune/Sharpless (epoxidation),2 Danishefsky (Diels-
Alder),3 J ohnson/Hudlicky (enzymatic desymmetrization),4
and Wong/Sharpless (osmium/enzyme),5 there still does not
exist a practical, nonenzymatic route to the hexoses.6 As part
of our program aimed at investigating the biological role of
oligosaccharides, we are interested in the synthesis of
analogues of the anti-HIV agent kakelokelose (Scheme 1).
Kakelokelose is a polysulfated â-oligomannosugar, isolated
as an oligomer from the Pacific tunicate Didemnum molle.
It has been proposed that the biologically active form is an
oligosaccharide.7 We are particularly interested in studying
oligosaccharide analogues of the all D-, all L-, and mixed D,L-
oligosugar structures. Consequently, we required an efficient
approach to both D- and L-mannose. The ideal route would
allow for the synthesis of other mixed D- and L-oligosaccha-
rides, such as the D-mannose-L-gulose portion of bleomycin.8
Herein, we would like to present our discovery of an
expeditious route to mannose, gulose, or talose using Sharp-
less’s dihydroxylation reaction to set the D- or L-configuration
depending on the ligand used.
We have targeted differentially protected D- and L-
mannose 1 as building blocks for the assembly of analogues
of kakelokelose (Scheme 1). We desired a route in which the
synthetic efficiency is better than traditional protection/
deprotection strategies from mannose and is amenable to
both D- and L-mannose. The ideal synthesis should also start
from a commercially available starting material that is even
cheaper than D-mannose.9,10 The strategy outlined below has
led to a highly stereocontrolled synthesis of D- or L-mannose
in only five steps and in approximately 39% yield from
furfural (Scheme 2). Our approach relies upon the use of
catalytic asymmetric osmium chemistry on vinylfuran 3
developed by Ogasawara11 and augments the earlier work
of Achmatowicz.12
To accomplish this goal we, as have others, recognized
that substituted furyl alcohols possess the proper C-5
stereochemistry for the hexopyranoses.13 Crucial to this
approach is a simple three-step route toward pyranone 6b
(Scheme 3). We envision 6b as the linchpin molecule that
will be amenable for the synthesis of all the possible
stereoisomers of the hexoses. A key part of this sequence is
our ability to convert furfural into an ether solution of
vinylfuran.14 This one-step in situ process for the generation
of vinylfuran constitutes a significant improvement in terms
of overall efficiency. A 2 M solution of vinylfuran can be used
directly in the t-BuOH/H2O AD-mix reaction mixture de-
veloped by Sharpless.12,15 The (DHQ)2Phal ligand gave an
85% yield of (R) diol 2(R) from furfural in 90% ee.16 The
(DHQD)2Phal ligand afforded (S) diol 2(S) in an 85% yield
with a 92% ee.17 The absolute configuration is based upon
the Sharpless mnemonic15 and Mosher ester analysis.18 Diol
2 can be selectively protected with TBSCl (90%) to give furan
5, which smoothly rearranges to hemiacetal 6a when
oxidized with NBS.19 Hemiacetal 6a exists as an equilibrat-
ing mixture of anomers, diastereomerically enriched mix-
tures of 6a equilibrate to a (1:1) mixture of anomers in
deuterated chloroform. Our hopes of taking advantage of the
difference in reactivities of axial versus equatorial anomeric
alcohols were realized when hemiacetal 6a was treated with
benzoyl chloride at -78 °C to produce pyranone 6b (>20:1
ratio of diastereomers). This result appears to be general
for acid chlorides as pivaloyl chloride (>10:1) showed similar
selectivity, whereas TBSCl gave approximately a 1:1 mixture
(11) (a) Achmatowicz, O.; Bielski, R. Carbohydr. Res. 1977, 55, 165. (b)
Grapsas, I.; K.; Couladouros, E. A.; Georgiadis, M. P. Pol. J . Chem. 1990,
64, 823.
(1) For a good review: Zamoiski, A.; Banaszek, A.; Grynkiewicz, G. Adv.
Carbohydr. Chem. Biochem. 1982, 40, 1.
(2) Sharpless, K. B.; Masamune, S. Science 1983, 220, 949.
(3) For a review, see: (a) Danishefsky, S. J . Chemtracts 1989, 273. For
improved catalysis, see: (b) Schaus, S. E.; Branalt, J .; J acobsen, E. N. J .
Org. Chem. 1998, 63, 403-405.
(4) (a) J ohnson, C. R.; Golebiowski, A.; Steensma, D. H.; Scialdone, M.
A. J . Org. Chem. 1993, 58, 7185-7194. (b) Hudlicky, T.; Pitzer, K. K.;
Stabile, M. R.; Thorpe, A. J .; Whited, G. M. J . Org. Chem. 1996, 61, 4151-
4153.
(12) Ogasawara has previously demonstrated the asymmetric dihydroxy-
lation of vinylfuran and applied it toward the synthesis of D- and L-
levoglucosenone: (a) Taniguchi, T.; Nakamura, K.; Ogasawara, K. Synlett
1996, 971. (b) Taniguchi, T.; Ohnishi, H.; Ogasawara, K. Chem. Commun.
1996, 1477-1478.
(13) For a good review, see: Hudlicky, T.; Entwistle, D. A.; Thorpe, A. J .
Chem. Rev. 1996, 96, 1195.
(14) This in situ use of vinylfuran allows for much higher yields of diol
7. Wittig technology provides vinylfuran in yields on the order of 10%.
Previous practical approaches to vinylfuran involve a four-step sequence
from furfural that involves a stoichiometric Cu-promoted decarboxylation
of 3-furylpropanoic acid, see: Schmidt, U.; Werner, J . Synthesis 1986, 986.
(15) Kolb, H. C.; VanNieuwenhze, M. S.; Sharpless, K. B. Chem. Rev.
1994, 94, 2483-2547.
(16) Although we were concerned about an adverse solvent effect caused
by the presence of ether, Ogasawara has observed identical ee’s in t-BuOH/
H2O; see ref 12a.
(5) Henderson, I.; Sharpless, K. B.; Wong, C.-H. J . Am. Chem. Soc. 1994,
116, 558-561.
(6) Recently Wong has developed a route to a mixture of glucose and
fructose using an enzymatic process: Gijsen, H. J . M.; Qiao, L.; Fitz, W.;
Wong, C.-H. Chem. Rev. 1996, 96, 443-73.
(7) Riccio, R.; Kinnel, R. B.; Bifulco, G.; Scheur, P. J . Tetrahedron Lett.
1996, 37, 1979.
(8) For a recent total synthesis of bleomycin, see: (a) Boger, D. L.; Honda,
T. J . Am. Chem. Soc. 1994, 116, 5647-5656. (b) Boger, D. L.; Honda, T.;
Menezes, R. F.; Colletti, S. L. J . Am. Chem. Soc. 1994, 116, 5631-5646. (c)
Boger, D. L.; Teramoto, S.; Zhou, J . C. J . Am. Chem. Soc. 1995, 117, 7344-
7356.
(17) The (R) diol 7 has been prepared on a half mole scale with no
reduction of enantioselectivity. Similarly the (S) diol 7 has been prepared
on a quarter mole scale.
(9) According to the Aldrich catalog, on a per gram basis D-mannose cost
35 times more than furfural and for L-mannose the cost ratio is 4000.
(10) Furfural is commercially produced by Great Lakes Chemical. For a
convenient laboratory procedure, see: Adams, R.; Voorhees, V. Organic
Syntheses; Wiley: New York, 1921; Collect. Vol. I, p 280.
(18) (a) Sullivan, G. R.; Dale, J . A.; Mosher, H. S. J . Org. Chem. 1975,
38, 2143. (b) Yamaguchi, S.; Yasuhara, F.; Kabuto, K. T. Tetrahedron 1976,
32, 1363.
(19) Georgiadis, M. P.; Couladoures, E. A. J . Org. Chem. 1986, 51, 2725
and ref 11b.
10.1021/jo990410+ CCC: $18.00 © 1999 American Chemical Society
Published on Web 04/08/1999