Development of a Novel Sugar Linkage: 6,6′-Ether-Connected Sugars

Article abstract of DOI:10.1002/anie.200352388

Unusual sugar dimers: The synthesis of 6,6′-ether-connected pyranoses by an acetalization-reduction procedure is described (see scheme). Structure-activity-relationship studies of these novel carbohydrate dimers suggested that the natural product coyolosa, which has been shown to have a significant effect on fasting blood-glucose levels, may have a different structure from the one previously reported.

Full text of DOI:10.1002/anie.200352388

Angewandte
Chemie
Dipyranose Ethers
Development of a Novel Sugar Linkage: 6,6’-
Ether-Connected Sugars**
Hideyo Takahashi, Toshimitsu Fukuda,
Haruhiko Mitsuzuka, Rie Namme, Hidetoshi Miyamoto,
Yasufumi Ohkura, and Shiro Ikegami*
Scheme 1. Synthesis of 6,6’-ether-connected pyranoses by the Williamson
etherification. DMF=N,N-dimethylformamide, Tos=p-toluenesulfonyl.
In 1997, PØrez et al. reported the structure 1 for coyolosa, a
natural product isolated from the root of Acrocomia mex-
icana[1] that was shown to have a
shown in Scheme 2. In this method, a 6-aldehyde 5 is treated
with a 6-alcohol 2 under mild acid-catalyzed conditions to
provide an intermediate acetal derivative 6. The treatment of
significant effect on fasting blood-glu-
cose levels.[2] Coyolosa may well be a
new candidate in the search for drugs
to combat diabetes. Upon examination
of the NMR spectroscopic data of
PØrez et al. for this unique carbohy-
drate in which two pyranose groups
are connected as a 6,6’-ether, we found
it unlikely that coyolosa had the pro-
posed structure 1.[3] Although there
can be no doubt that two pyranose rings linked through their
6-positions form the structure of coyolosa,[4] it is less evident
that the stereochemistry of the pyranose moieties corre-
sponds to that shown in 1. Herein, we report a novel synthesis
of 6,6’-ether-connected pyranoses and propose an absolute
configuration for coyolosa based on structure–activity-rela-
tionship (SAR) studies that is different to the one originally
suggested.
Little has been reported to date on sugars linked by ether
bonds. In preliminary studies, we treated the methyl a-d-
gluco-pyranoside 2a with the iodide and the tosylate corre-
sponding to 3 under basic conditions in an attempted
Williamson etherification. The desired ether 4a was only
obtained in low to moderate yields (Scheme 1), as the
anomeric position was unstable under the Williamson con-
ditions, and the competing elimination could not be avoided.
As the Williamson etherification had proved unsuccessful
in our attempt to synthesize 6,6’-ether-connected pyranoses,
we instead adopted an acetalization–reduction approach, as
Scheme 2. Proposed synthetic routes to 6,6’-ether-connected pyranoses 4
(shown for 4a).
6 with a reducing agent then furnishes the desired product 4.
As a two-step procedure is inconvenient, a one-pot reductive
etherification,[5] which is a more reliable method for the
preparation of ethers under nonbasic conditions, was also
envisaged.
A large number of studies have been carried out on
reductive etherification, that is, the reduction of oxocarbe-
nium ions or acetals generated in situ to provide ethers. In
particular, Lewis acid catalyzed reactions of these compounds
with alkoxy silanes or hydrosilanes have proved useful for the
efficient synthesis of unsymmetrical ethers.[5e–j] Although the
reaction of simple carbonyl compounds (e.g. benzaldehyde)
with various alcohols has already been reported, there are few
methods[4j] suitable for carbonyl compounds substituted with
bulky or multifunctional carbohydrates.
[*] Prof. S. Ikegami, Dr. H. Takahashi, T. Fukuda, H. Mitsuzuka,
R. Namme
Acetal formation was investigated first. An equimolar
amount of the methyl a-d-gluco-pyranoside 2a reacted with
the aldehyde 5a very slowly to give the acetal derivative 6a in
poor yield. However, when an excess of 2a was used in the
School of Pharmaceutical Sciences, Teikyo University
Sagamiko, Kanagawa 199-0195 (Japan)
Fax: (+81)426-85-3729
E-mail: shi-ike@pharm.teikyo-u.ac.jp
Dr. H. Miyamoto, Dr. Y. Ohkura
presence
of
trimethylsilyl
trifluoromethanesulfonate
Department of Pharmacology, Research Laboratories
KOTOBUKI Pharmaceutical Co., Ltd.
6351 Sakaki, Nagano 389-0697 (Japan)
(TMSOTf) at 08C, 6a was isolated in 91% yield. It had
been anticipated that the acetal derivative, which contains
three bulky pyranosides, would be relatively unstable. How-
ever, we found that 6a was so stable that it could be purified
without difficulty by silica-gel column chromatography. The
structure of this acetal derivative is unusual and has many
features that remain to be investigated.[6]
[**] We thank J. Shimode and M. Kitsukawa for spectroscopic mea-
surements. Partial financial support for this research from the
Ministry of Education, Culture, Sports, Science, and Technology is
gratefully acknowledged. Support from the Takeda Science Foun-
dation to H.T. is also acknowledged.
The reactions of the d-gluco-, d-galacto-, d-manno-, and
d-allo-pyranosides 2a, 2b, 2c , and 2d, respectively, with the
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2003, 42, 5069 –5071
DOI: 10.1002/anie.200352388
ꢀ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5069

Communications
aldehyde derivatives 5a–d[7] under similar conditions afforded
the corresponding acetal derivatives 6a–j (Table 1). The
excess of the pyranoside 2 was recovered quantitatively in
each case and reused in other reactions.
of the synthesis of these types of carbohydrates, which are
connected through the 6,6’-positions by an ether linkage
rather than by the usual glycoside linkage.[8]
Having established the required acetalization–reduction
procedures, we next turned our attention to the development
of a more convenient, one-pot synthesis. First, the one-pot
etherification of 2a with 5a was investigated. Their acetaliza-
tion in the presence of TMSOTf was carried out at 08C for
2 h, and subsequent addition of TMSOTf and TESH to this
system led to the conversion of the acetal into 4a in 89% yield
(Scheme 3). However, in this case an excess of 2a and the
other reagents was required for the reaction to reach
completion. To establish a more efficient synthetic method,
we examined the possibility of decreasing the amount of 2a
required by substituting it in the reaction with its 6-OTMS
derivative 7a. After surveying a variety of conditions, we
found that 3 equivalents of 7a were sufficient to provide 4a in
92% yield (Scheme 4). The 6-OTMS group may play a role in
decreasing the degree of coordination between 7a and
TMSOTf or TESH.
Table 1: Acetalization of aldehydes 5 with alcohols 2 catalyzed by
TMSOTf.[a]
Entry
5
2 (6-OH)
6
Yield [%][b]
1
2
3
4
5
6
7
8
9
a (Glc)
a (Glc)
a (Glc)
b (Gal)
b (Gal)
b (Gal)
c (Man)
c (Man)
c (Man)
d (All)
a (Glc)
b (Gal)
c (Man)
a (Glc)
b (Gal)
c (Man)
a (Glc)
b (Gal)
c (Man)
d (All)
a (Glc-Glc-Glc)91
b (Gal-Glc-Gal)81
c (Man-Glc-Man)93
d (Glc-Gal-Glc)93
e (Gal-Gal-Gal)76
f (Man-Gal-Man)84
g (Glc-Man-Glc)75
h (Gal-Man-Gal)76
i (Man-Man-Man)80
j (All-All-All)69
10
[a] Reactions were conducted with 10 equivalents of 2. [b] Yield of
isolated product.
With the desired intermediates 6 in hand, we then
investigated the reduction step with triethylsilane (TESH)
as the reducing agent. It was found that TMSOTf accelerated
the reduction of 6 with TESH efficiently to give the desired
6,6’-ether-connected pyranosides 4. The acetals 6a–j were
successfully converted into the corresponding pyranosides
4a–g and the excess of the alcohol 2 could be recovered
quantitatively and reused (Table 2). This is the first example
Scheme 3. One-pot acetalization–reduction reaction of 2a and 5a with
triethylsilane in the presence of trimethylsilyl trifluoromethanesulfonate.
Table 2: Reduction of acetals 6 with triethylsilane catalyzed by TMSOTf.
Scheme 4. One-pot acetalization–reduction of 5a and 7a with triethyl-
silane in the presence of trimethylsilyl trifluoromethanesulfonate.
TMS=trimethylsilyl.
Entry
6
4
Yield [%][a]
To complete the synthesis of coyolosa and its analogues
we still needed to unmask the hydroxy groups in 4. Removal
of the benzyl groups of 4a–g by hydrogenolysis furnished the
corresponding pyranosides 8a–g. Treatment of 8a–g with
trifluoroacetic acid then gave the desired 6,6’-ether-connected
pyranoses 9a–g in moderate to good yields (Table 3).
1
2
3
4
5
6
7
8
9
a (Glc-Glc-Glc)
b (Gal-Glc-Gal)
c (Man-Glc-Man)
d (Glc-Gal-Glc)
e (Gal-Gal-Gal)
f (Man-Gal-Man)
g (Glc-Man-Glc)
h (Gal-Man-Gal)
i (Man-Man-Man)
j (All-All-All)
a (Glc-Glc)96
b (Glc-Gal)79
c (Glc-Man)95
b (Gal-Glc)75
d (Gal-Gal)82
e (Gal-Man)73
c (Man-Glc)73
e (Man-Gal)69
f (Man-Man)64
g (All-All)56
We inspected the spectral data[9] of 9a–g for confirmation
of their structure and reconsidered the possibility of the
proposed structure 1 of coyolosa based on structure–activity-
relationship (SAR) studies. According to the procedure
reported,[1] the biological activity of all synthetic 6,6’-ether-
10
[a] Yield of isolated product.
5070
ꢀ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2003, 42, 5069 –5071

Angewandte
Chemie
7a (70.7 mg, 0.13 mmol) in CH2Cl2 (0.2 mL) at À788C. The reaction
Table 3: Synthesis of coyolosa and analogues.
mixture was stirred at À208C for 70 min, then triethylsilane (70 mL,
0.44 mmol) was added at À208C. The reaction mixture was stirred at
08C for 4 h then poured into a saturated solution of NaHCO3. The
mixture was extracted with CH2Cl2, and the organic phase was dried
over Na2SO4 and concentrated in vacuo. The residual oil was purified
by silica-gel column chromatography (toluene/AcOEt 2:1), and the
excess of 2a was recovered quantitatively. Silica-gel column-chromat-
ographic purification (hexane/AcOEt = 4:1) of the remaining mate-
rial afforded the desired product 4a (36.8 mg, 0.042 mmol, 92%) as a
colorless solid.
Entry
4
9
Yield [%][a]
1
2
3
4
5
6
7
a (Glc-Glc)
b (Glc-Gal)
c (Glc-Man)
d (Gal-Gal)
e (Gal-Man)
f (Man-Man)
g (All-All)
a
b
c
d
e
f
78
50
72
59
76
37
59
Received: July 18, 2003 [Z52388]
Published Online: October 6, 2003
Keywords: antidiabetic agents · carbohydrates · ethers ·
.
structure–activity relationships · synthetic methods
g
[a] Yield from 4.
[1] S. PØrez, G. R. M. PØrez, G. C. PØrez, G. M. A. Zavala, S. R.
Vargas, Pharm. Acta Helv. 1997, 72, 105 – 111.
[2] It was reported that coyolosa showed a significant blood-sugar-
lowering effect on normal and alloxane-induced diabetic mice
when administered at doses of 2.5–40 mgkgÀ1 i.p. (intraperito-
neal); see reference [1].
[3] It was reported by PØrez et al. that the NMR spectrum of
coyolosa was measured as a solution in CDCl3; however, it is
impossible to dissolve unprotected pyranoses in CDCl3. There-
fore, we believe there may be an error in the description given in
the report. We were unsuccessful in our attempts to contact the
authors and were therefore unable to obtain a sample of the
natural product from them.
connected pyranoses 9a–g on glucose tolerance in alloxane-
induced diabetic rats was investigated. Surprisingly, 9 f
demonstrated the most favorable effect on fasting blood-
glucose levels, which was similar to that of coyolosa, as
reported by PØrez et al.[1,10] We therefore tentatively propose
that the structure of coyolosa might be the mannose-derived
isomer 9 f, if this natural product is indeed a 6,6’-ether-
connected carbohydrate.
In conclusion, we have developed a novel synthesis of 6,6’-
ether-connected pyranoses through an acetalization–reduc-
tion procedure. For convenience, a one-pot synthesis was also
established. Based on an SAR study, we suggest that the
structure of coyolosa might be the 6,6’-ether-connected
mannose. Additional synthetic and biological studies on
ether-connected carbohydrates are currently underway in
our laboratory.
[4] The structure was determined mainly by mass spectrometry,
which is a reliable medium.
[5] a) M. P. Doyle, D. J. Debruyn, S. J. Donnelly, D. A. Kooistra,
A. A. Odubela, C. T. West, S. M. Zonnebelt, J. Org. Chem. 1974,
39, 2740; b) M. P. Doyle, C. T. West, S. J. Donnelly, C. C.
McOsker, J. Organomet. Chem. 1976, 117, 129; c) M. B. Sassa-
man, G. K. Prakash, G. A. Olah, Tetrahedron 1988, 44, 3771;
d) M. P. Doyle, D. J. Debruyn, D. A. Kooistra, J. Am. Chem. Soc.
1972, 94, 3659; e) K. C. Nicolaou, C.-K. Hwang, D. A. Nugiel, J.
Am. Chem. Soc. 1989, 111, 4136; f) J.-I. Kato, N. Iwasawa, T.
Mukaiyama, Chem. Lett. 1985, 743 – 746; g) S. Hatakeyama, H.
Mori, K. Kitano, H. Yamada, M. Nishizawa, Tetrahedron Lett.
1994, 35, 4367 – 4370; h) K. Miura, K. Ootsuka, S. Suda, H.
Nishikiori, A. Hosomi, Synlett 2002, 313 – 315; i) X. Jiang, J. S.
Bajwa, J. Slade, K. Prasad, O. Repic, T. J. Blacklock, Tetrahedron
Lett. 2002, 43, 9225 – 9227; j) C.-C. Wang, J.-C. Lee, S.-Y. Luo, H.-
F. Fan, C.-L. Pai, W.-C. Yang, L.-D. Lu, S.-C. Hung, Angew.
Chem. 2002, 114, 2466 – 2468; Angew. Chem. Int. Ed. 2002, 41,
2360 – 2362; k) T. Suzuki, K. Ohashi, T. Oriyama, Synthesis 1999,
1561 – 1563; l) M. Wada, S. Nagayama, K. Mizutani, R. Hiroi, N.
Miyoshi, Chem. Lett. 2002, 248 – 249; m) S. H. Lee, Y. J. Park,
C. M. Yoon, Tetrahedron Lett. 1999, 40, 6049 – 6050; n) M. J.
Verhoef, E. J. Creyghton, J. A. Peters, H. V. Bekkum, Chem.
Commun. 1997, 1989 – 1990.
Experimental Section
General procedure for acetalization: TMSOTf (46 mL, 0.25 mmol)
was added to a mixture of 5a (58.9 mg, 0.13 mmol) and 2a (592 mg,
1.30 mmol) in CH2Cl2 (2.5 mL) at 08C. The reaction mixture was
stirred for 2 h then poured into a saturated solution of NaHCO3. The
mixture was extracted with CH2Cl2, and the organic phase was dried
over Na2SO4 and concentrated in vacuo. The residual oil was
subjected to silica-gel column chromatography (toluene/AcOEt
2:1), and the excess of 2a was recovered quantitatively. Silica-gel
column-chromatographic purification (hexane/Et2O 1:1) of the
remaining material afforded 6a (159.7 mg, 0.12 mmol, 91%) as an oil.
General procedure for acetal reduction: TMSOTf (21 mL,
0.12 mmol) and triethylsilane (123 mL, 0.77 mmol) were added
successively to a solution of 6a (52.8 mg, 0.038 mmol) in CH2Cl2
(0.4 mL) at À788C. The reaction mixture was stirred at À208C for 4 h
then poured into a saturated solution of NaHCO3. The mixture was
extracted with CH2Cl2, and the organic phase was dried over Na2SO4
and concentrated in vacuo. The residual oil was subjected to silica-gel
column chromatography (toluene/AcOEt 2:1), and the excess of 2a
was recovered quantitatively. Silica-gel column-chromatographic
purification (hexane/AcOEt 4:1) of the remaining material afforded
the desired product 4a (33.5 mg, 0.037 mmol, 96%) as a colorless
solid.
[6] Further investigations are in progress.
[7] A. J. Mancuso, D. Swern, Synthesis 1981, 165 – 168.
[8] No nomenclature has been established as yet by the IUPAC for
these types of pyranoses connected by ether bonds. Hence, we
hesitate to assign them names.
[9] Spectral (1H NMR, 13C NMR) and analytical (HRMS) data
showed 9a–g to be the 6,6’-ether-connected pyranoses. See
Supporting Information.
[10] A dosage of 60 mgkgÀ1 of 9 f caused a maximum blood-sugar
lowering of 26% in 2 h in alloxane-induced diabetic rats.
One-pot procedure for the synthesis of 4: TMSOTf (79 mL,
0.44 mmol) was added to a solution of 5a (20.3 mg, 0.044 mmol) and
Angew. Chem. Int. Ed. 2003, 42, 5069 –5071
www.angewandte.org
ꢀ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5071