Convenient, in situ generation of anhydrous hydrogen iodide for the preparation of α-glycosyl iodides and vicinal lodohydrins and for the catalysis of Ferrier glycosylation

Article abstract of DOI:10.1021/ol991312d

(matrix presented) Anhydrous hydrogen iodide is generated in situ by the reaction of solid iodine and a thiol. The HI thus generated has been employed for the efficient preparation of α-glycosyl iodides and vicinal iodohydrins from the corresponding glycosyl acetates and epoxides, respectively, and for Ferrier glycosylation of alcohols and thiols.

Full text of DOI:10.1021/ol991312d

ORGANIC
LETTERS
2000
Vol. 2, No. 3
369-372
Convenient, in Situ Generation of
Anhydrous Hydrogen Iodide for the
Preparation of r-Glycosyl Iodides and
Vicinal Iodohydrins and for the
Catalysis of Ferrier Glycosylation
Stephanie M. Chervin, Paolo Abada, and Masato Koreeda*
Department of Chemistry, UniVersity of Michigan, Ann Arbor, Michigan 48109-1055
koreeda@umich.edu
Received December 6, 1999
ABSTRACT
Anhydrous hydrogen iodide is generated in situ by the reaction of solid iodine and a thiol. The HI thus generated has been employed for the
efficient preparation of r-glycosyl iodides and vicinal iodohydrins from the corresponding glycosyl acetates and epoxides, respectively, and
for Ferrier glycosylation of alcohols and thiols.
In connection with our interest in the design and synthesis
of sialyl Lewis X mimetics, the use of the per-O-acyl lactosyl
halide 1 was required as a glycosyl donor. Given the
Gervay procedure requires the use of the expensive and
highly labile reagent trimethylsilyl iodide.3 The adaptation
of the widely used method for the preparation of per-O-
acetylated glycosyl bromides4 (which entails treatment of the
per-O-acetylated glycosides with anhydrous HBr in acetic
acid, frequently at reflux) for the preparation of the corre-
sponding iodides does not appear practical, due to the
difficulty in obtaining anhydrous HI.5
(2) (a) Ness, R. K.; Fletcher, H. G.; Hudson, C. S. J. Am. Chem. Soc.
1951, 73, 296-300. (b) Krontzer, F. K.; Schuerch, C. Carbohydr. Res. 1974,
34, 71-78. (c) Ferrari, B.; Pavia, A. A. Carbohydr. Res. 1980, 79, C1-
C7. (d) Thiem, J.; Meyer, B. Chem. Ber. 1980, 113, 3075-3085. (e) Klemer,
A.; Bieber, M. Liebigs Ann. Chem. 1984, 1052-1055. (f) Ernst, B.; Winkler,
T. Tetrahedron Lett. 1989, 30, 3081-3084. (g) Lichtenthaler, F. W.; Kla¨res,
U.; Lergenmu¨ller, M.; Schwidetzky, S. Synthesis 1992, 179-184. (h)
Paulsen, H.; Rutz, V.; Brockhausen, I. Liebigs Ann. Chem. 1992, 747-
758. (i) Premchandran, R. H.; Ogletree, M. L.; Fried, J. J. Org. Chem. 1993,
58, 5724-5731. (j) Lichtenthaler, F. W.; Ku¨hler, B. Carbohydr. Res. 1994,
258, 77-85. (k) Uchiyama, T.; Hindsgaul, O. Synlett 1996, 499-501. (l)
Schmid, U.; Waldemann, H. Tetrahedron Lett. 1996, 37, 3837-3840. (m)
Schmid, U.; Waldemann, H. Liebigs Ann./Recl. 1997, 2573-2577.
(3) (a) Gervay, J.; Hadd, M. J. J. Org. Chem. 1997, 62, 6961-6967. (b)
Gervay, J.; Nguyen, T. N.; Hadd, M. J. Carbohydr. Res. 1997, 300, 119-
125.
enhanced reactivity of the lactosyl iodide 1b toward nucleo-
philic displacement over the lactosyl bromide 1a,1 we were
particularly interested in the efficient preparation of 1b.
While a number of reliable methods to access glycosyl
iodides have been reported,2 Gervay’s recently published
preparation appears to be the only one that allows for the
isolation of the stereochemically pure, reportedly unstable
iodides in a relatively efficient manner.3 However, even this
(1) (a) Hashimoto, S.; Honda, T.; Ikegami, S. Tetrahedron Lett. 1990,
31, 4769-4772. (b) Schmid, U.; Waldemann, H. Tetrahedron Lett. 1996,
37, 3837-3840.
(4) See, e.g.: Helferich, B.; Weis, K. Chem. Ber. 1956, 89, 314-318.
10.1021/ol991312d CCC: $19.00 © 2000 American Chemical Society
Published on Web 01/06/2000

1
Concurrently, in our investigations of the iodine-catalyzed
Ferrier reaction,6 it was observed that the reaction of tri-O-
acetyl glucal and a thiol catalyzed by iodine (20 mol %) in
THF resulted in the formation of the byproduct 4-iodobutan-
1-ol,7 apparently the reaction product of the solvent molecule
with in situ generated HI. Investigation into this reagent
mixture indicated that the source of the HI was likely the
result of the oxidation of the thiol in the presence of iodine.
Therefore, it was envisaged that the reaction of iodine with
a thiol in an organic solvent might provide a synthetically
useful source of anhydrous HI. HI thus generated could be
conveniently applied to the synthesis of glycosyl iodides from
per-O-acetylated glycosyl acetates and the synthetically
useful vicinal iodohydrins from the corresponding epoxides
and could serve as an efficient catalyst for the Ferrier
glycosylation reaction.
monitored by H NMR spectroscopy in CDCl3. Both thio-
lacetic acid (4) and 1,3-propanedithiol (6) were oxidized
quantitatively to their respective disulfides, and the HI adduct
with 2-methyl-2-butene (2), 2-iodo-2-methylbutane (3), was
produced cleanly (Table 1). Notably, an NMR signal
Table 1. Addition of HI to Alkene 2a Monitored by H NMR
1
Spectroscopy
The oxidation of thiols with iodine is the basis of well-
established methods for the iodometric titration of thiol
groups,8 for the preparation of peptide disulfide linkages,9
and for the oligomerization of dithiols.10,11 However, it should
be noted that, in all reported cases, the oxidation proceeds
to a significant extent only if the HI byproduct is removed
by reaction with a base or by solvation in a biphasic reaction
mixture. In the general scheme (Scheme 1), a thiol molecule
aReaction conditions: 2 (0.245 mmol), 4 or 6 (0.245 mmol), and I2 (0.123
mmol) were dissolved at room temperature in CDCl3 (1 mL) in a 5 mm
1
o.d. NMR tube. The reaction was monitored by H NMR spectroscopy at
1
400 MHz. b No H NMR signals corresponding to alkene 2 detected.
assignable to HI was not detected in either case. As expected,
Scheme 1
the same reaction of 4 or 6 in the absence of butene 2 did
1
not produce any disulfide product, as judged by H NMR
spectroscopy.
Treatment of the per-O-acetyl derivatives of mono- and
disaccharides with a solution of iodine (0.5 mol equiv) and
thiolacetic acid (4; 1.0 mol equiv) in dichloromethane at
reflux resulted in the rapid, highly stereoselective formation
of the corresponding R-glycosyl iodides (Table 2). The crude
reaction mixtures were amenable to silica gel flash column
chromatography, and the products were isolated as white
amorphous solids. Comparison of the reactivity of per-O-
acetyl-â-D-glucose (entry 1) and per-O-acetyl-R-D-glucose
(entry 2) reveals that the efficiency of the conversion is
dependent upon the relative geometry of the anomeric acetate
group and the neighboring acetate group at C-2. Presumably,
in the 1,2-trans compounds, the acetate group at C-2 is likely
to assist ionization of the anomeric leaving group, leading
directly to a stabilized oxonium intermediate. In contrast, in
the 1,2-cis systems, such ionization is not possible and a
much lower yield is observed. This type of C-2 acetate group
participation is well-documented for various cases of ano-
meric group displacement.12 The exclusive formation of the
R-iodides regardless of the configuration of the starting
anomeric acetate group is a manifestation of the rapid
equilibrium of the initially produced iodides that favors the
thermodynamically more stable R-iodides.3b
reacts with molecular iodine to form an intermediate sulfenyl
iodo species and one molecule of HI. The sulfenyl iodide
reacts with the second thiol molecule to yield the corre-
sponding disulfide and a second molecule of HI. While this
method has been widely recognized as a route to the disulfide
species, little attention has been given to its potential for
the stoichiometric production of anhydrous hydrogen iodide.
In an effort to assess the efficiency for the formation of
HI as the byproduct of the oxidation of thiols, the reaction
of 1.0 mol equiv of thiols 4 and 6 with iodine (0.5 mol equiv)
in the presence of 2-methyl-2-butene (2; 1 mol equiv) was
(5) For the generation of anhydrous HI, see the following. (a) I2/
tetrahydronaphthalene: Hoffman, C. J. Inorg. Synth. 1963, 7, 180-182.
(b) I(py)2BF4/CH2Cl2 + HBF4/Et2O; Et3SiH: Barluenga, J.; Gonza´lez, J.
M.; Campos, P. J.; Asensio, G. Angew. Chem., Int. Ed. Engl. 1985, 24,
319-320. (c) I2 + activated Al2O3/petroleum ether: Pagni, R. M.; Kabalka,
G. W.; Boothe, R.; Gaetano, K.; Stewart, L. J.; Conaway, R.; Dial, C.;
Gray, D.; Larson, S.; Luidhardt, T. J. Org. Chem. 1988, 53, 4477-4482.
(d) Et2PhN‚BI3/AcOH: Reddy, Ch. K.; Periasamy, M. Tetrahedron Lett.
1990, 31, 1919-1920. See also: (e) Irifune, S.; Kibayashi, T.; Ishii, Y.;
Ogawa, M. Synthesis 1988, 366-369. (f) Brommfield, C. E. Org. Proc.
Res. DeV. 1997, 1, 88-89.
(6) Koreeda, M.; Houston, T. A.; Shull, B. K.; Klemke, E.; Tuinman,
R. J. Synlett 1995, 90-92.
(7) Tuinman, R. J.; Koreeda, M. Unpublished results.
(8) Capozzi, G.; Modena, G. In The Chemistry of the Thiol Group; Patai,
S., Ed.; Wiley: London, 1974; Part 2, pp 785-839.
In view of the 1′,2′-trans geometry at the interglycosidic
linkage of lactose, the possible cleavage of the interglycosidic
(9) Farlow, M. W. J. Biol. Chem. 1948, 176, 71-72.
(10) Musker, W. K.; Goodrow, M. H.; Olmstead, M. M. Phosphorus
Sulfur Relat. Elem. 1983, 16, 299-302.
(11) Amaratunga, W.; Milne, J.; Santagati, A. J. Polym. Sci. A: Polym.
Chem. 1998, 36, 379-390.
(12) Collins, P.; Ferrier, R. Monosaccharides: The Chemistry and Their
Roles in Natural Products; Wiley: New York, 1995; pp 60-188.
370
Org. Lett., Vol. 2, No. 3, 2000

Scheme 2
Table 2. Synthesis of R-Glycosyl Iodides from Per-O-acetyl
Sugars with in Situ Generated HI
nucleophilic capture of the oxonium ion intermediate or the
direct displacement of the R-glycosyl iodide product.
Vicinal iodohydrins are highly versatile intermediates in
synthesis.13,14 While numerous mild methods to access such
reactive intermediates exist, stereo- and regiocontrolled
opening of epoxides to vicinal halohydrins continues to
attract attention from synthetic chemists.15-18 In this context,
transformation of epoxides into vicinal halohydrins was
examined with the use of the present in situ generated HI.
As summarized in Table 3, treatment of various epoxides
a δ(H-1) and J1,2 in CDCl3 (400 MHz): 10 (7.00, 4.0 Hz); 12 (6.71, 0.7
Hz); 1b (6.92, 6.4 Hz); 16 (9.60, 4.4 Hz). b Yield of isolated, chromato-
graphically pure products.
Table 3. Ring Opening of Epoxides with in Situ Generated
HIa
bond by HI was a concern. Thus, treatment of per-O-acetyl-
â-D-lactose (13) with 1 equiv of HI at longer reaction times
(>2 h) resulted in the formation of 2,3,4,6-tetra-O-acetyl-
R-galactosyl iodide, albeit in low yield (<10%), and in the
diminished yield of the lactosyl iodide 1b. Interestingly, more
vigorous, but brief, treatment with the reagent led to the
efficient formation of the lactosyl iodide 1b. Thus, treatment
of per-O-acetyl-â-D-lactose with 2 equiv of thiolacetic acid
and 1 equiv of iodine for 20 min in refluxing dichlo-
romethane afforded, upon purification, the desired R-glycosyl
iodide 1b in 77% yield. No products of interglycosidic bond
cleavage were detected in the cases of per-O-acetyl maltose
substrates (entries 5 and 6), reflecting the lower susceptibility
toward hydrolysis of the 1′,2′-cis geometry of the maltose
system.
Although the efficient formation of glycosyl iodides could
be achieved, with the use of the in situ generated HI, by
starting from per-O-acetylated carbohydrates, those substrates
such as per-N,O-acetylated 2-deoxyamino and 2-deoxy
sugars did not yield their corresponding anomeric iodides
even with the use of excess iodine and thiolacetic acid. While
the reasons for this apparent lack of reactivity in these sugars
remain unclear, in the case of 2-amino sugars, it is likely
that protonation at the more basic 2-acetylamide oxygen
electronically disfavors the formation of the intermediate
oxonium ion or its equivalent. Although both thiols 4 and 6
proved efficient for the production of HI from iodine for
addition to an alkene as described above, the use of 1,3-
propanedithiol for the generation of HI for the synthesis of
glycosyl iodides led to the production of the â-thioglycoside
(17; Scheme 2). This thioglycoside may result from the
a Reaction conditions: epoxide compound (1 mol equiv) stirred with I2
(0.5 mol equiv) and thiolacetic acid (4; 1 mol equiv) in CH2Cl2 at room
temperature for 20 min. b Yield of isolated, chromatographically pure
products.
with the HI resulted in the stereospecific formation of the
corresponding vicinal iodohydrins in nearly quantitative yield
with the exception of the case of styrene oxide (27). The
(13) (a) Maruda, H.; Takase, K.; Nishio, M.; Hasegawa, A.; Nishiyama,
Y.; Ishii, Y. J. Org. Chem. 1994, 58, 5550-5555 and references therein.
(b) Horiuchi, C. A.; Ikeda, A.; Kanamori, M.; Hosokawa, H.; Sugiyama,
T.; Takahashi, T. T. J. Chem. Res., Synop. 1997, 60-61.
Org. Lett., Vol. 2, No. 3, 2000
371

ring opening of styrene oxide was shown not to be highly
regioselective (entry 5). In addition, the optical purity of the
styrene oxide regenerated from iodohydrin 28 (NaOMe/
MeOH) was found to have eroded from 98 to only 35%,
suggesting significant involvement of the SN1 pathway for
the formation of 28 from styrene oxide.
Table 4. HI-Catalyzed Glycosylation of Alcohols and Thiols
As part of our continuing interest in developing new
methods for the glycosylation of alcohols, phenols, and other
pharmacologically important molecules, it was discovered
that the Ferrier-type glycosylation can take place in a highly
efficient manner in the presence of catalytic iodine.6 While
the exact mechanism of this unique reaction remains to be
elucidated, preliminary evidence suggests that the glycosy-
lation proceeds as a result of the production of small amounts
of HI from the iodine and substrate. Therefore, it was felt
that anhydrous HI produced by the oxidation of a thiol with
iodine might also be an effective catalyst for the Ferrier-
glycosylation reaction.19 In this regard, the glycoside donor
tri-O-acetyl glucal was treated with various acceptors in the
presence of catalytic amounts of HI (Table 4). It was found
that glycosylated products 32 and 33 were obtained stereo-
selectively in quantitative yield in less than 30 min with 5
mol % of the catalyst with simple glycosyl acceptors (entries
1 and 2). A 1,6-linked disaccharide, 37, was obtained in
excellent yield (80%, purified) by this method with good
anomeric stereoselectivity (entry 5). Glycosylation of thiol
acceptors, while practical in a few cases (see entries 3 and
4), are less efficient in certain cases due to the competing
reaction of the thiol with iodine.
a Method A: 5 mol % of HI generated from I2 and thiolacetic acid (4)
in CH2Cl2. Method B: 5 mol % of HI generated from I2 and 1,3-
propanedithiol (6) in CH2Cl2. b Yield of isolated, chromatographically pure
products. c Ratio of the two anomers at the newly created glycosylic centers.
synthesis of R-glycosyl iodides and vicinal iodohydrins from
their corresponding per-O-acetyl carbohydrates and epoxides,
respectively (both with stoichiometric HI) and for catalysis
of the Ferrier glycosylation of alcohols and thiols (with
catalytic HI). The byproduct disulfides 5 and 7 (see Table
1) have shown no adverse effect on the outcome of these
reactions. Further applications on the use of this conveniently
generated anhydrous HI in synthesis are currently under
investigation.
In summary, we have shown that anhydrous HI (1 mol
equiv) in dichloromethane can be generated quantitatively
from the oxidation of a thiol compound (1 mol equiv) by
solid iodine (0.5 mol equiv) in the presence of a substrate
that reacts with the in situ generated HI. The HI thus
generated has been employed for the highly stereocontrolled
(14) Larock, R. C. ComprehensiVe Organic Transformations; VCH: New
York, 1989; pp 508-509.
(15) Stewart, C. A.; VanderWerf, C. A. J. Am. Chem. Soc. 1954, 76,
1259-1264.
Supporting Information Available: Experimental pro-
cedures and characterization data for the new compounds
described. This material is available free of charge via the
Internet at http://pubs.acs.org.
(16) Vidari, G.; Garlaschelli, L. Gazz. Chim. Ital. 1987, 117, 251-253.
(17) Bajwa, J. S.; Anderson, R. C. Tetrahedron Lett. 1991, 32, 3021-
3024.
(18) Konaklieva, M. I.; Dahl, M. L.; Turos, E. Tetrahedron Lett. 1992,
33, 7093-7096.
(19) Ferrier, R. J.; Prasad, N. J. Chem. Soc. C 1969, 570-575.
OL991312D
372
Org. Lett., Vol. 2, No. 3, 2000