A convenient large-scale synthesis of methyl α-maltoside: A simple model for amylose

Article abstract of DOI:10.1016/S0008-6215(98)00110-4

Methyl 4-O-(α-D-glucopyranosyl)-α-D-glucopyranoside (methyl α-maltoside), a model compound for amylose, has been synthesized in four steps and 63% overall yield from relatively inexpensive D-(+)-maltose.

Full text of DOI:10.1016/S0008-6215(98)00110-4

Carbohydrate Research 308 (1998) 345±348
A convenient large-scale synthesis of methyl
ꢀ-maltoside: a simple model for amylose
Stuart J. Gebbiea, Ian Gosneya,*, Paul R. Harrisonb,
Isabelle M.F. Lacana, William R. Sandersonb, J. Phillip Sankeyb
a Department of Chemistry, The University of Edinburgh, West Mains Road, Edinburgh, UK, EH9 3JJ
b Solvay Interox Research and Development, PO Box 51, Moor®eld Road, Widnes, UK, WA8 0FE
Received 31 October 1997; accepted 17 April 1998
Abstract
Methyl 4-O-(ꢀ-d-glucopyranosyl)-ꢀ-d-glucopyranoside (methyl ꢀ-maltoside), a model com-
pound for amylose, has been synthesized in four steps and 63% overall yield from relatively inex-
pensive d-(+)-maltose. # 1998 Elsevier Science Ltd. All rights reserved
Keywords: Methyl ꢀ-maltoside; Amylose model; Stereoselective synthesis
1. Introduction
in the literature. Of these, two routes use methyl ꢀ-
d-glucopyranoside as the starting material for
enzymatic transformations performed respectively
by Dextrin glycotransferase in conjunction with
amylopectin [3], and an unidenti®ed pentylase
enzyme isolated from Bacillus macerans with
cyclohexaamylose [4]. An eight-step convergent
synthetic route has also been used to construct 5ꢀ,
albeit in 31% overall yield (with no yield being
given for two individual steps) [5]. Finally, Dick et
al. [6] have also reported the synthesis of 5ꢀ by a
slightly shorter route based on d-(+)-maltose, but
it is marred by repeated acetylation/de-acetylation
steps and the necessity for both selective removal
of admixed ꢁ-anomer by oxidation with chromic
acid and large-scale chromatography. Herein we
describe a much more ecient and practical route
for the large-scale synthesis of 5ꢀ from d-(+)-mal-
tose 1. This is accomplished in a straightforward
manner as shown in Scheme 1 without recourse to
the use of either an enzymatic step or the need for
column chromatography.
Methyl 4-O-methyl-ꢀ-d-glucopyranoside has
long been used as a model compound for the key
polysaccharide amylose [1] in a variety of studies
[2] and whilst it possesses the necessary structural
features in the correct stereochemical arrangement,
a considerable drawback is that the ether linkage is
much more resistant to hydrolysis than the unit
linkages in the natural product. A more complex,
and consequently a more representative model,
should contain at least one glucose-glucose linkage
and also be protected at the terminal anomeric
position since in the natural compound the major-
ity of glucose units are protected at this position by
other glucose units. The simplest compound to
meet these criteria is methyl 4-O-(ꢀ-d-glucopyr-
anosyl)-ꢀ-d-glucopyranoside (methyl ꢀ-maltoside)
5ꢀ for which four syntheses have appeared hitherto
* Corresponding author. Fax: 0131-650-4743.
0008-6215/98/$19.00 # 1998 Elsevier Science Ltd. All rights reserved
PII S0008-6215(98)00110-4

346
S.J. Gebbie et al./Carbohydrate Research 308 (1998) 345±348
Scheme 1. Reagents and conditions: (i) 10.6 eq. pyridine/10.6 eq. acetic anhydride/0.02 eq. N,N-dimethylaminopyridine/CH2Cl2/
31 h/0 ꢀC!room temperature; (ii) NH3(sat)/MeOH:THF (3:7)/100 min/room temperature; (iii) 1.2 eq. Ag2O/1.2 eq. MeI/MeCN/
60 min/ca. 80 ꢀC: or alternatively, 2.0 eq. Ag2O/2.0 eq. MeI/MeCN/72 h/room temperature; (iv) 3.5 eq. NaOH/MeOH/10 min/room
temperature.
2. Results and Discussion
3. Experimental
The ®rst step involving peracetylation of a rig-
orously dried sample of maltose 1 was achieved by
reaction with an excess of freshly distilled acetic
anhydride in the presence of N,N-dimethylamino-
pyridine [7]. Next, the resulting ꢁ-maltose octaace-
tate 2ꢁ was subjected to selective anomeric
deacetylation under conditions which are equally
e€ective on milligram and hundred gram scale [8].
Thus, 2ꢁ was stirred at low temperature in a mixed
solvent (3 parts MeOH:7 parts THF) saturated
with ammonia [9]. After ca. 100 min, the resulting
anomeric mixture of aldoses 3 was crystallised
selectively, or alternatively used without further
puri®cation in the subsequent methylation step
accomplished [10] via a modi®cation of the
traditional Purdie approach. In the absence of
light, methyl iodide was added to a rapidly stirred
suspension of 3 (either as an anomeric mixture or
anomerically pure form) and silver (I) oxide in
acetonitrile. Upon heating in a Soveril-joint
ampoule for 60 min, an anomeric mixture of 4 (ca.
Thin-layer chromatography (TLC) was carried
out on silica gel 60 F254 plates and visualised by
dipping the plate into a solution of concentrated
1
sulfuric acid in ethanol (5:95). H- and 13C-NMR
spectra were obtained on a Bruker AC-250 spec-
trometer operating at 250 and 62.9 MHz, respec-
tively. All other physical data including optical
rotations which were measured on an Optical
Activity AA 1000 polarimeter at 589 nm (the
sodium D-line) using a 1 dm cell, unless quoted,
were in accord with previously reported data.
Maltose octaacetate (2b).ÐTo a rapidly stirring
suspension of a rigorously dried sample of 1
(19.10 g, 56 mmol) in dichloromethane (150 mL)
and pyridine (47.8 mL, 591 mmol) was added N,N-
dimethylaminopyridine (0.14 , 1.1 mmol) under
ꢀ
argon at 0 C. After 30 min, freshly distilled acetic
anhydride (55.8 mL, 591 mmol) was added slowly
to the reaction mixture and when TLC (ethyl acet-
ate:cyclohexane 4:1) had indicated the formation of
one major fraction and no remaining starting
material, it was quenched by addition of hydro-
chloric acid (2 M, 200 mL). After separation and
extraction with dichloromethane (3Â25 mL), the
combined organic layers were washed with satu-
rated aqueous sodium hydrogen carbonate
(200 mL), then saturated aqueous sodium chloride
(200 mL) and dried over magnesium sulfate before
1
8:1 ꢀ:ꢁ (by H NMR spectroscopy) was obtained
in 81% yield. If on the other hand, the reaction
mixture was allowed to stir at ambient tempera-
ture in a sealed tube for three days and isolated
from light, the ꢀ-anomer of 4 was obtained exclu-
sively in 83% yield, presumably as a consequence
of the thermodynamic equilibrium in favour of 3ꢀ.
In the last step, complete anomeric deprotection of
4ꢀ was achieved quantitatively by stirring in
methanol with wet sodium hydroxide to furnish 5ꢀ
in 63% overall yield and complete stereochemical
integrity. An added bene®t of this shorter route
was the eciency of all of the individual steps
irrespective of whether the reactions were carried
out on a small (ca. 100 mg) or large (ca. 100 g)
scale.
evaporation in vacuo to give an amorphous col-
20
ourless solid, [ꢀŠ ‡ 65ꢀꢁc ˆ 1, ethanol), lit. [11].
d
+64 ꢀ(34.84 g, 92%) of sucient purity to use
without further puri®cation in subsequent steps.
1H NMR (CDCl3): ꢂH 5.68 (d, 1H, J1,2 8.0, H-1),
5.30 (d, 1H, J;1 ,2 4.0, H-10), 5.28±5.20 (cm, 2H, H-
0
0
3, H-30), 5.01 (t, 1H, J3 ,4
,
4 ,5 ,10.0, H-4 ), 4.91 (d0d,
J
0
0
0
0
0
0
0
1H, J2,3 9.0, H-2), 4.80 (dd, 1H, J2 ,3 , 10.0, H-2 ),
4.37 (dd, 1H, J5,6a 2.0 J5,6b 5.0, H-6a), 4.24±4.13

S.J. Gebbie et al./Carbohydrate Research 308 (1998) 345±348
347
(cm, 2H, H-6b, H-6a,), 4.03±3.95 (cm, 2H, H-4, H-
6b,), 3.90±3.85 (cm, 1H, H-5,), 3.81±3.75 (cm, 1H,
H-5), 2.07±1.93 (cm, 8ÂCOCH3), 13C NMR
(CDCl3): ꢂC 170.3±168.5 (8ÂCOCH3), 95.5, (CH),
91.0, (CH), 75.0 (CH), 72.8 (CH), 72.3 (CH), 70.7,
(CH), 69.8 (CH), 69.1 (CH), 68.4 (CH), 67.8 (CH),
62.3 (CH2), 61.2 (CH2), 20.8±20.2 (8ÂCOCH3).
Maltose 2,3,6,20,30,40,50-heptaacetate (3a and
3b).Ðꢁ-Maltose octaacetate 2ꢁ, 20.00 g, 29.5mmol)
(7 mL) was added silver oxide (0.44 g, 1.9 mmol) and
the mixture was stirred for 5 min. with the vessel
protected from light. After this time, iodomethane
(0.15 mL, 1.9 mmol, freshly ®ltered through a plug
of neutral aluminium oxide) was added, theꢀvessel
sealed and immersed in an oil bath at ca. 90 C for
90 min. Upon cooling to room temperature, the
reaction mixture was diluted with dichloromethane
(20 mL), ®ltered through a plug of celite and evap-
orated in vacuo to give a faintly yellow crystalline
foam which was recrystallised from ethanol as a
colourless amorphous solid (0.86 g, 81%, ꢀ : ꢁ 8:1
by 1H NMR spectroscopy). ꢀ-form 1H NMR
(CDCl3): ꢂH 5.35 (d, 1H, J1,2 4.0, H-1), 5.30±5.11
ꢀ
was added at 78 C to a mixed solvent (metha-
nol:tetrahydrofuran 3:7, 450 mL) through which
ammonia had been bubbled vigorously over a
25 min period. After stirring for 10 min, the reaction
was allowed to warm to room temperature. When
TLC (ethyl acetate:cyclohexane 4:1) had indicated
the complete consumption of the starting material
and the formation of a single, more polar fraction,
the solvent and excess ammonia were removed by
evaporation in vacuo to yield a pale yellow syrup
which solidi®ed on standing to a glass. Although
not essential, this glass may be recrystallised slowly
from ethanol:water (9:1) to provide 3ꢀ solely, or as
an anomeric mixture of 3ꢀ and 3ꢁ from iso-propa-
nol to yield an amorphous solid (15.61 g, 82%). ꢀ-
(cm, 2H, H-3, H-30), 4.99 (t, 1H, J3 ,4 10.0, J4 ,5
0
0
0
0
10.0, H-40), 4.82 (d, 1H, J1 ,2 4.0, H-10), 4.75 (dd,
1H, J2,3 9.0, H-2), 4.41 (cm, 1H, H-20), 4.30±4.16
(cm, 3H, H-5, H-6a0, H6b0), 4.01±3.75 (cm, 3H, H-4,
H-6a0, H-6b0), 3.66±3.57 (cm, 1H, H-50), 3.43 (s, 3H,
OCH3), 2.09±1.94 (cm, 21H, 7ÂCOCH3). 13C NMR
(CDCl3): ꢂC 170.3±169.2 (7ÂCOCH3), 100.9 (CH),
95.3 (CH), 75.2 (CH), 72.5 (CH), 71.9 (2ÂCH), 69.8
(CH), 69.1 (CH), 68.3 (CH), 67.8 (CH), 62.6 (CH2),
61.3 (CH2), 56.8 (OCH3), 20.7±20.4 (7ÂCOCH3).
Methyl a-d-heptaacetyl-maltoside (4a).ÐTo an
anomeric mixture of aldose 3 (1.01 g, 1.6 mmol)
suspended in acetonitrile (7 mL) in the dark, was
added silver oxide (0.77 g, 3.3 mmol) with stirring.
After 15 min, iodomethane (0.21 mL, 3.3 mmol,
0
0
1
form H NMR (CDCl3): ꢂH 5.52 (dd, 1H, J2,3 9.0,
0
0
J3,4 10.0, H-3), 5.37 (d, 1H, J1 ,2 , 4.0, H-1 ), 5.30 (d,
0
0
0
0
0
0
1H, J1,2 4.0, H-1), 5.22 (t, 1H, J2 ,3 , 9.0, J3 ,4 , 9.0,
H-30), 5.00 (dd, 1H, J4 ,5 , 10.0, H-4 ), 4.70 (dd, 2H,
J2,3 10.0, H-2, H-2,), 4.43 (dd, 1H, J5,6a 3.0, J6a,6b
13.0, H-6a), 4.26-4.14 (cm, 3H, H-5, H-6a, H-6b),
3.99, (dd, 1H, J5 ,6a , 3.5, J6a ,6b , 12.0, H-6a ), 3.99±
3.89 (cm, 2H, H-4, H-50), 2.08±1.94 (21H, 7ÂCH3);
13C NMR (CDCl3): ꢂC 170.5±169.3 (7ÂCOCH3),
95.3 (CH), 89.7 (CH), 72.5 (CH), 72.1 (CH), 71.4
(CH), 69.8 (CH), 69.2 (CH), 68.2 (CH), 67.8 (CH),
67.4 (CH), 62.6 (CH2), 61.2 (CH2), 20.7±20.3
0
0
freshly ®ltered through
a
plug of neutral
0
0
0
0
0
aluminium oxide) was added and the mixture left
to stir over a 72 h period. Afterwards, the reaction
mixture was diluted with dichloromethane (20 mL),
®ltered through a plug of celite, and evaporated in
vacuo to a crystalline foam which was recrys-
tallised from iso-propanol to yield 4ꢀ (vide supra)
as colourless needles (0.86 g, 83%).
Methyl a-d-maltoside (5a).ÐTo a rapidly stirred
solution of 4ꢀ (4.53 g, 7.0 mmol) in methanol
(25 mL) was added freshly ground sodium hydro-
xide (0.98 g, 24.4 mmol). After 10 min, TLC (ethyl
acetate) showed complete consumption of starting
material and the presence of only one very polar
fraction. The reaction mixture was neutralised by
the addition of Dowex MR-3 ion exchange resin,
and evaporated in vacuo to a crystalline foam
[ꢀ]d20+176 ꢀ(c=1, water), lit. + 174 ꢀ [6] (2.48 g,
100%). 1H NMR (D2O): ꢂH 5.20 (d, 1H, J1,2 4.0,
1
(7ÂCOCH3). ꢁ-form H NMR (CDCl3): ꢂH 5.37
(d, 1H, J1 ,2 4.0, H-10), 5.30 (t, 1H, J2,3 9.0, J3,4
0
0
9.0, H-3), 5.22 (t, 1H, J2 ,3 9.0, J3 ,4 9.0, H-30), 5.00
0
0
0
0
0
0
0
0
(dd, 1H, J4 ,5 10.0, H-4 ), 4.81 (dd, 1H, H-2 ), 4.72±
4.65 (cm, 2H, H-1, H-2), 4.43 (dd, 1H, J5,6a 3.0,
J6a,6b 13.0, H-6a), 4.26±4.14 (cm, 2H, H-6b0, H-
6a0), 3.99 (dd, 1H, J5 ,6a 3.5, J6a ,6b 12.5, H-6a0),
3.99±3.89 (cm, 2H, H-50, H-4), 3.75±3.66 (cm, 1H,
H-5), 2.08-1.95 (21H, 7ÂCOCH3). 13C NMR
(CDCl3): ꢂC 170.5±169.3 (7ÂCOCH3), 95.3 (CH),
94.6 (CH), 74.8 (CH), 73.5 (CH), 72.5 (CH), 69.8
(CH), 69.2 (CH), 68.2 (CH), 67.8 (CH), 67.4 (CH),
62.6 (CH2), 61.2 (CH2), 20.7±20.3 (7ÂCOCH3).
Methyl a/b-d-heptaacetyl-maltoside (4a and
0
0
0
0
H-1), 4.29 (d, 1H, J1 ,2 8.0, H-10), 4.10±3.22 (cm,
12H, ring protons), 3.46 (s, 3H, OCH3); 13C NMR
(CD3OD): ꢂC 101.4 (CH), 100.1 (CH), 78.5 (CH),
74.8 (CH), 74.0 (CH), 73.6 (CH), 72.7 (CH), 72.0
0
0
4b).ÐTo
a
Soveril-joint ampoule containing
aldose 3 (1.01g, 1.6 mmol) suspended in acetonitrile

348
S.J. Gebbie et al./Carbohydrate Research 308 (1998) 345±348
(CH), 71.3 (CH), 70.0 (CH), 61.8 (CH2), 61.6
(CH2), 55.6 (OCH3).
Res., 273 (1995) 249±253 for practical syntheses of
methyl 4-O-methyl-ꢀ-d-glucopyranoside.
[3] S. Peat, W.J. Whelan, and G. Jones, J. Chem. Soc.,
(1957) 2490±2495.
[4] J.H. Pazur, J.M. Marsh and T. Ando, J. Am.
Chem. Soc., 81 (1959) 2170±2172.
[5] I. Backman, B. Erbing, P-E. Jansson, and L.
Kenne, J. Chem. Soc., Perkin Trans. I, (1988) 889±
898 and references cited therein.
Acknowledgements
We appreciate greatly the use of the EPSRC
funded Chemical Database Service at Daresbury.
[6] W.E. Dick Jr., D. Weisleder, and J.E. Hodge,
Carbohydr. Res, 18 (1971) 115±123.
[7] G. Ho¯e, W. Steglich, and H. Vorbruggen, Angew.
Chem. Int. Ed. Engl., 17 (1978) 569±583.
[8] See G.-T. Ong, K.-Y. Chang, S.-H. Wu, and K.-T.
Wang, Carbohydr. Res., 265 (1994) 311±318 for an
enzymatic procedure but limited to ca. 10 g of
compound.
[9] X-X. Zhu, P.Y. Ding, and M-S. Cai, Tetrahedron
Asymm., 7 (1996) 2833±2838.
[10] N. Finch, J.J. Fitt, and I.H.S. Hsu, J. Org. Chem.,
40 (1975) 206±215.
References
[1] For a general review of the industrial importance,
see R.L. Whistler and J.R. Daniel, Starch, in M.
Howe-Green, J.J. Kroschwitz, and D.F. Othmer
(Eds.), Kirk±Othmer Encyclopedia of Chemical
Technology, 4th ed., Vol. 22, Wiley, New York,
1991, pp 699±719 and references cited therein.
[11] N.K. Kochetkov, E.M. Klimov, and N.M.
Malysheva, Tetrahedron Lett., 30, (1989), 5459±
5462.
[2] See inter alia R.L. Whistler and S.J. Kazeniac, J.
Am. Chem. Soc., 76 (1954) 3044-3045; and more
recently L. Kaichang and R.F. Helm, Carbohydr.