2-Deoxy-2-iodo-beta-glucopyranosyl fluorides: mild and highly stereoselective glycosyl donors for the synthesis of 2-deoxy-beta-glycosides from beta-hydroxy ketones.

Article abstract of DOI:10.1021/ol027257h

2-Deoxy-2-iodo-beta-glucopyranosyl fluoride 14 is a highly stereoselective glucopyranosyl donor that may be activated under mild conditions. Application of this new glycosyl donor to the glycosidation reactions of a variety of acceptors including beta-hyd

Full text of DOI:10.1021/ol027257h

ORGANIC
LETTERS
2003
Vol. 5, No. 1
81-84
2-Deoxy-2-iodo-â-glucopyranosyl
Fluorides: Mild and Highly
Stereoselective Glycosyl Donors for the
Synthesis of 2-Deoxy-â-glycosides from
â-Hydroxy Ketones
Nicolas Blanchard and William R. Roush*
Department of Chemistry, UniVersity of Michigan, 930 North UniVersity,
Ann Arbor, Michigan 48109-1055.
roush@umich.edu
Received November 9, 2002
ABSTRACT
2-Deoxy-2-iodo-â-glucopyranosyl fluoride 14 is a highly stereoselective glucopyranosyl donor that may be activated under mild conditions.
Application of this new glycosyl donor to the glycosidation reactions of a variety of acceptors including â-hydroxy ketones affords â-glycosides
with high efficiency and stereoselectivity.
2-Deoxy-glycosides are important structural units found in
numerous natural and biologically active compounds such
as the angucycline family of antibiotics (landomycin A), the
aureolic acid antibiotics (olivomycin A, chromomycin A3),
2-iodo-â-glucopyranosyl acetates 14 and 2-deoxy-2-iodo-R-
glucopyranosyl trichloroacetimidates 25 are highly reactive
glycosyl donors for establishing â-linked glycosides. The
C(2)-iodo unit can then be reductively removed6 under mild
I
the enediynes (calicheamycin γ1 , esperamicins A1 and C),
the avermectins (avermectin B1a, ivermectin), some choles-
tane glycosides (OSW-1), and cardiac glycosides.1 Although
some general methods have been developed for the stereo-
selective construction of 2-deoxy-R-glycosidic linkages
(mainly by electrophilic addition to glycals),2 preparation of
the corresponding â-linkage has proved to be much more
difficult. Our group has been involved in the development
of new methods of synthesis of this challenging 2-deoxy-
â-glycosidic linkage.3-5 We previously reported that 2-deoxy-
(2) (a) Veyrie`res, A. Special Problems in Glycosidation Reactions:
2-Deoxy Sugars. In Carbohydrates in Chemistry and Biology; Ernst, B.,
Hart, G. W., Sinay¨, P., Eds.; Wiley-VCH: Weinheim, Germany, 2000; Part
I, Vol. I, p 367. (b) Marzabadi, C. H.; Franck, R. W. Tetrahedron 2000,
56, 8385. (c) Kirshning, A.; Bechthold, A. F.-W.; Rohr, J. Top. Curr. Chem.
1997, 188, 1. (d) Danishefsky, S. J.; Bilodeau, M. T. Angew. Chem., Int.
Ed. Engl. 1996, 35, 1380. (e) Toshima, K.; Tatsuta, K. Chem. ReV. 1993,
93, 1503.
(3) Leading references: (a) Roush, W. R.; Hartz, R. A.; Gustin, D. J. J.
Am. Chem. Soc. 1999, 121, 1990. (b) Roush, W. R.; Sebesta, D. P.; Bennett,
C. E. Tetrahedron 1997, 53, 8825, 8837. (c) Roush, W. R.; Briner, K.;
Kesler, B. S.; Murphy, M.; Gustin, D. J. J. Org. Chem. 1996, 61, 6098.
(4) Roush, W. R.; Bennett, C. E. J. Am. Chem. Soc. 1999, 121, 3541.
(5) (a) Roush, W. R.; Chong, P. Org. Lett. 2002, 4, 4523. (b) Roush, W.
R.; Gung, B. W.; Bennett, C. E. Org. Lett. 1999, 1, 891.
(1) (a) Albrecht, H. P. Cardiac Glycosides. In Naturally Occurring
Glycosides; Ikan, R., Ed.; Wiley: Chichester, U.K., 1999; p 83. (b)
Weymouth-Wilson, A. C. Nat. Prod. Rep. 1997, 14, 99.
(6) Miura, K.; Ichinose, Y.; Nozaki, K.; Fugami, K.; Oshima, K.;
Utimoto, K. Bull. Chem. Soc. Jpn. 1989, 62, 143.
10.1021/ol027257h CCC: $25.00 © 2003 American Chemical Society
Published on Web 12/12/2002

not yet available. An intrinsic limitation in the glycosidation
of â-hydroxy ketones is the weakened reactivity of the
acceptor, due to intramolecular hydrogen bonding of the
hydroxyl hydrogen to the carbonyl moiety.8 Selected relevant
examples of â-selective glycosidation reactions of aldol
acceptors with 2-deoxy donors have been reported by Tastuta
and Kinoshita.9 Glycosidation of a â-hydroxy ketone with a
2-deoxy glycosyl fluoride under modified Mukaiyama’s
conditions10 proceeded in 30% yield and unreported anomeric
selectivity. Upon reinvestigation of this reaction in connection
with a total synthesis of concanamycin A, Paterson11 obtained
the glycoside product in only 12% yield and a â:R ratio of
1:1.4. After screening a variety of 2-deoxyglycosyl donors,
Paterson determined that the 2-deoxy-glycosyl bromide was
the most â-selective of those examined. The desired 2-deoxy-
â-glycoside was obtained with 2.5:1 â:R selectivity in 21%
yield.11 In Evans’ total synthesis of cytovaricin,12 the
glycosidation of a â-hydroxy Weinreb amide derivative with
a 2-deoxy glycosyl acetate using trityl perchlorate activation
provided the â-glycoside product in 70% yield with a â: R
selectivity of 4:1. This glycosidation reaction was extremely
sensitive to variation of protecting groups, solvents, and
temperatures.
Scheme 1. Glycosidation Reaction of
2-Iodo-2-deoxy-glycoside Acetates 1 and Trichloroacetimidates
2
conditions, leading to the desired 2-deoxy-â-glycosidic unit
in high yield (Scheme 1).
During the course of our studies directed toward the total
synthesis of formamicin (Figure 1),7 a complex member of
the plecomacrolide family of antibiotics, we anticipated the
need to perform a â-selective glycosidation reaction of a late-
stage â-hydroxy ketone (aldol) intermediate. We report herein
the results of our studies of this key glycosidation reaction
using model substrates, which led to the development of a
highly selective synthesis of 2-deoxy-â-glycosides using
2-deoxy-2-iodo-â-glycosyl fluorides as the glycosyl donors.
The lack of general methods available to efficiently
glycosidate aldols in a â-selective manifold presented a
unique opportunity to test our highly reactive and stereo-
selective 2-deoxy-2-iodo-glucopyranosyl acetate (1) and
trichloroacetimidate (2) methodology.
Glycosidation reactions of â-hydroxy ketone 12 (corre-
sponding to the C18-C24 fragment of formamicin) with
2-iodo-2-deoxy glycosyl acetates (5-7), imidates (8-10),
or phosphate13 11 in the presence of a variety of promoters
(TMSOTf, BF3‚OEt2, TrClO4,12 K10 clay,14 LiClO4,15
LiOTf16) led only to decomposition of the acceptor (Scheme
2). Control experiments showed that in the presence of 0.3
equiv of TMSOTf, the acceptor 12 was not stable for more
than 20 min at low temperature (-78 to -30 °C). Optimiza-
tion of the glycosidation with TBSOTf (0.3 equiv) as the
promoter led to a disappointing 21% yield of the desired
glycosylated product 13 but with excellent anomeric selectiv-
Figure 1. Formamicin and the targeted late stage â-hydroxy ketone
(aldol) glycosidation substrate.
Scheme 2. Glycosidation Reaction of â-Hydroxy Ketone 12
Although â-selective glycosidation reactions of â-hydroxy
carbonyl compounds are known in the literature,9,11,12
a
general method proceeding in good yield and selectivity is
(7) (a) Igarashi, M.; Kinoshita, N.; Ikeda, T.; Nakagawa, E.; Hamada,
M.; Takeuchi, T. J. Antibiot. 1997, 50, 926. (b) Igarashi, M.; Nakamura,
H.; Naganawa, H.; Takeuchi, T. J. Antibiot. 1997, 50, 932. Correction: J.
Antibiot. 1998, 51, C1.
(8) A similar hydrogen-bonding pattern has been reported to increase
the nucleophilicity of hydrogen-bound hydroxyl acceptors: Mitchell, S. A.;
Pratt, M. R.; Hruby, V. J.; Polt, R. J. Org. Chem. 2001, 66, 2327. However,
our data indicate that hydrogen-bound aldols such as 12 and 16 are
considerably less reactive than acceptors 19 and 21.
(9) Toshima, K.; Misawa, M.; Ohta, K.; Tatsuta, K.; Kinoshita, M.
Tetrahedron Lett. 1989, 30, 6417.
(10) Mukaiyama, T.; Murai, Y.; Shoda, S.-i. Chem. Lett. 1981, 431.
(11) Paterson, I.; McLeod, M. D. Tetrahedron Lett. 1995, 36, 9065.
(12) Evans, D. A.; Kaldor, S. W.; Jones, T. K.; Clardy, J.; Stout, T. J. J.
Am. Chem. Soc. 1990, 112, 7001.
82
Org. Lett., Vol. 5, No. 1, 2003

ity (â:R > 98: 2).17 This example further confirms that
TBSOTf is superior to TMSOTf for the glycosidation
reactions of sensitive substrates.2b,5,18 However, to achieve
a synthetically useful glycosidation procedure with aldol
acceptors, it was clear that almost neutral activation condi-
tions of the donor would be required.19
of hemiacetals to the glycosyl fluoride 14 by using DAST23
proceeded in 78% yield and excellent anomeric stereoselec-
tivity (â:R > 98: 2). The configuration of the anomeric center
was determined by measurement of the coupling constant
(J1-2 ) 8.4 Hz)17 after desilylation (HF‚NEt3, CH3CN, 60
°C).24 Donor 14 is relatively stable and could be stored at
-20 °C for more than two weeks without any noticeable
decomposition.
We anticipated that a donor combining a C(2)-iodo
substituent, which we have shown to be a very efficient
â-directing group,4,5 with an anomeric fluoride leaving group
might help to circumvent the stability problems noted above.
2-Iodo-2-deoxy glycosyl fluorides were first reported by
Wood et al. in 1966,20 and have continued to be targets of
methodological studies for the past 30 years.21 However, only
one report of glycosidation reactions involving this class of
donor has been disclosed, in which cyclohexanol was used
as the acceptor.22 Therefore, we decided to explore the
potential of 2-iodo-2-deoxy-â-glycosyl fluorides in the
glycosidation reactions of â-hydroxy ketones.
With the glycosyl donor 14 in hand, we turned our
attention toward the glycosidation reaction of the â-hydroxy
ketone 12. To our delight, slow addition of donor 14 to a
solution of â-hydroxy ketone 12, stannous chloride, and silver
perchlorate in diethyl ether at -15 °C according to
Mukaiyama’s general procedure10 provided the coupled
product 13 in 65% yield with excellent stereoselectivity (â:R
> 98: 2)17 (Scheme 4). Silver triflate proved equally effective
2-Iodo-2-deoxy-glycosyl fluorides are easily prepared in
two steps starting from the corresponding 2-iodo-2-deoxy-
â-glycosyl acetate (Scheme 3). For the present purposes, we
Scheme 4. Glycosidation Reactions of
2-Iodo-2-deoxy-â-glycosyl Fluoride 14 and Aldols 12 and 16
Scheme 3. Synthesis of 2-Iodo-2-deoxy-â-glycosyl Fluoride
14
elected to use the readily accessible glycosyl acetate 54 as
starting material. Use of a substrate with a C(6)-bromo
substituent simplifies the overall synthetic sequence, in that
the C(2)-iodo and C(6)-bromo substituents can be reduced
in the same step to give the targeted 2,6-dideoxy-â-glycosides
(vide infra). Thus, cleavage of the anomeric acetate unit of
5 with hydrazine followed by transformation of the mixture
as the activating agent (58% isolated yield of 13) whereas
addition of a base (2,6-lutidine)25 led to a lower yield (24%).
Other promoters (AgClO4/Cp2HfCl2,26 AgClO4/Cp2ZrCl2,27
AgSbF6/SnCl2) or incorporation of a more labile protecting
group on the aldol acceptor (TES instead of TBS ether)
resulted in unsuccessful glycosidation reactions. More elabo-
rated â-hydroxy ketones (e.g., 16) can also be glycosylated
with the 2-iodo-2-deoxy-â-glycosyl fluoride 14 in very good
yield and selectivity (84%, â:R > 98: 2).17 Reductive removal
of the C(2)-iodo and the C(6)-bromo substituents under mild
conditions6 led to the desired 2-deoxy-glycoside units 15 and
18 in 72% and 90%, respectively.
(13) Lee, J.; Coward, J. K. J. Org. Chem. 1992, 57, 4126.
(14) Nagai, H.; Matsumura, S.; Toshima, K. Tetrahedron Lett. 2002, 43,
847.
(15) Waldmann, H.; Bo¨hm, G.; Schmid, U.; Ro¨ttele, H. Angew. Chem.,
Int. Ed. Engl. 1994, 33, 1944.
(16) Lubineau, A.; Drouillat, B. J. Carbohydr. Chem. 1997, 16, 1179.
(17) Determined by 500-MHz 1H NMR analysis of the crude reaction
mixture.
(18) Roush, W. R.; Narayan, S. Org. Lett. 1999, 1, 899.
(19) Activation of a glycal donor in the presence of 12 was also
attempted, but none of the literature methods examined (e.g., PPh3‚HBr,
NBS) led to formation of the â-glycoside product. The acceptor was
recovered in these cases. PPh3‚HBr actiVation: Bolitt, V.; Mioskowski,
C.; Lee, S.-G.; Falck, J. R. J. Org. Chem. 1990, 55, 5812. NBS actiVation:
Toshima, K.; Tatsuta, K.; Kinoshita, M. Bull. Chem. Soc. Jpn. 1988, 61,
2369.
(20) Wood, K. R.; Kent, P. W.; Fisher, D. J. Chem. Soc. (C) 1966, 912.
(21) (a) Evans, R. D.; Schauble, J. H. Synthesis 1987, 551. (b) Campbell,
J. C.; Dwek, R. A.; Kent, P. W.; Prout, C. K. Carbohydr. Res. 1969, 10,
71. (c) Hall, L. D.; Manville, J. F. Can. J. Chem. 1969, 47, 361. (d) Hall,
L. D.; Manville, J. F. Carbohydr. Res. 1968, 8, 295. (e) Hall, L. D.; Manville,
J. F. Chem. Commun. 1968, 35, 37.
(23) (a) Rosenbrook, Wm., Jr.; Riley, D. A.; Lartey, P. A. Tetrahedron
Lett. 1985, 26, 3. (b) Posner, G. H.; Haines, S. R. Tetrahedron Lett. 1985,
26, 5.
(24) Determination of the anomeric configuration of 14 by measurement
of the J1,2 coupling constant was not possible since this donor exists in a
twist-boat conformation to relieve the gauche interactions between the two
bulky silyl ethers at C3 and C4.
(25) Nicolaou, K. C.; Caulfield, T.; Kataoka, H.; Kumazawa, T. J. Am.
Chem. Soc. 1988, 110, 7910.
(26) Suzuki, K.; Maeta, H.; Matsumoto, T.; Tsuchihashi, G.-i. Tetra-
hedron Lett. 1988, 29, 3571.
(22) Nishimura, S.; Washitani, K. (Sumitomo Pharmaceuticals Co., Ltd.,
Japan), Stereoselective Production of Glycosyl Compound, Japanese Patent
09241288, 1997.
(27) Matsumoto, T.; Maeta, H.; Suzuki, K.; Tsuchihashi, G.-i. Tetra-
hedron Lett. 1988, 29, 3567.
Org. Lett., Vol. 5, No. 1, 2003
83

To test the scope of this glycosylation procedure, we
subjected a variety of acceptors to glycosidation reactions
with 14. The results are presented in Scheme 5.
undergoes this challenging glycosidation and permits di-
saccharide 24 to be obtained in 89% yield, the â-anomer
being the only detectable diastereomer in this reaction.17
Increasing the sensitivity of the acceptor as in the case of
25 led to several products, the major arising from Ferrier
rearrangement.29 The rearranged coupled product 26 was
isolated in 40% yield.
Scheme 5. Glycosidation Reactions of Various Acceptors with
2-Iodo-2-deoxy-â-glycosyl Fluoride 14
It is interesting to note that the conditions for activation
of donors 1, 2, and 14 can be controlled such that it should
be possible to use 6-heteroatom-substituted 2-iodo-glycosyl
acetates or 2-iodo-glycosyl fluorides as acceptors in gly-
cosidation reactions with 2-iodo-glycosyl trichloroacet-
imidates as the donors. The differential reactivity of these
three classes of glycosyl donors should be of considerable
utility in the synthesis of oligosaccharides containing 2-deoxy-
glycoside units.
In summary, we have shown that 2-iodo-2-deoxy-â-
glycosyl fluoride 14 is a synthetically useful glycosyl donor
for establishing â-glycosidic linkages with a variety of
acceptors. In particular, donor 14 gave excellent results in
the glycosidation reaction of aldol acceptors 12 and 16.
Application of this methodology to the total synthesis of
formamicin is currently underway and will be reported in
due course.
Acknowledgment. We thank the National Institutes of
Health (GM 38436) for support of this research. N.B. also
acknowledges the Ministe`re Franc¸ais des Affaires Etrange`res
for a Lavoisier Fellowship. We also thank Dr. Brad Savall
for initial studies of the glycosidation reactions of aldol
acceptors.
We have previously shown that the primary and secondary
alcohols 19 and 21 are glycosylated in high yields and
â-selectivities with donors such as 5, under TMSOTf
activation.4 The 2-iodo-2-deoxy-â-glycosyl fluoride 14 proved
equally effective, with even greater selectivity in the case
of the acceptor 21 (â:R > 98: 2 compared to â:R g 90:10
for the experiment with 54). Acid-sensitive acceptors such
as 23 are not stable above -50 °C in the presence of a strong
Lewis acid such as TMSOTf.18 Consequently, the glycosi-
dation of 23 in the presence of catalytic amounts of TMSOTf
with the 2-iodo-2-deoxy-â-glycosyl acetate 5 proceeds in less
than 11% yield (â:R > 98: 2).28 Donor 14, however,
Supporting Information Available: Experimental pro-
cedures and spectral data for compounds 12-18, 24, and
26. This material is available free of charge via the Internet
at http://pubs.acs.org.
OL027257H
(29) We suspect that 26 arises via Ferrier-type decomposition of 25 to
give an equivalent of Lewis acid complexed benzyloxide, which then
undergoes standard Ferrier substitution with the cation generated from 25.
The resulting 3,4-unsaturated sugar then presumably undergoes subsequent,
slower, glycosidation with 14 to give the observed product, 26. It is of
course conceivable that the order of these steps could be reversed. For
additional examples of this process, see ref 28.
(28) Bennett, C. E. Ph.D. Thesis, Indiana University, Bloomington, IN,
2000.
84
Org. Lett., Vol. 5, No. 1, 2003