Efficient Transport of Saccharides through a Liquid Membrane Mediated by a Cyclodextrin Dimer

Article abstract of DOI:10.1002/chem.200304816

A new efficient system for transporting saccharides through a liquid membrane has been constructed. The transport rates of saccharides were accelerated greatly by the cyclodextrin dimer 2; by contrast, the corresponding cyclodextrin monomer 1 was not effective at mediating saccharide transport. The transport rate of D-ribose through a chloroform liquid membrane was 17 times faster when the cyclodextrin dimer 2 was used as the transporter than when the cyclodextrin monomer 1 was used. Similarly the transport rate of methyl D-galactopyranoside was 16 times faster by 2 than by 1.

Full text of DOI:10.1002/chem.200304816

FULL PAPER
Efficient Transport of Saccharides through a Liquid Membrane Mediated by a
Cyclodextrin Dimer
Hiroshi Ikeda,* Akiyuki Matsuhisa, and Akihiko Ueno[a]
Abstract: A new efficient system for transporting saccharides through a liquid
membrane has been constructed. The transport rates of saccharides were accelerated
greatly by the cyclodextrin dimer 2; by contrast, the corresponding cyclodextrin
monomer 1 was not effective at mediating saccharide transport. The transport rate of
d-ribose through a chloroform liquid membrane was 17 times faster when the
cyclodextrin dimer 2 was used as the transporter than when the cyclodextrin
monomer 1 was used. Similarly the transport rate of methyl d-galactopyranoside was
16 times faster by 2 than by 1.
Keywords:
cyclodextrins ¥ host guest systems
liquid membrane molecular
recognition
carbohydrates
¥
¥
¥
effective for the selective transport of saccharide.[5] Cyclo-
dextrins (CDs) are cyclic oligosaccharides composed of six,
seven, and eight glucose units joined by a-1,4-linkages named
Introduction
Recently, much effort has been devoted to the construction of
systems for effectively transporting saccharides through a
liquid membrane containing of artificial transporters, because
such systems are useful not only for separating mixtures of
saccharides but also for elucidating the mechanism underlying
8]
as a-, b-, and g-CD, respectively.[6 Because CD has many
hydroxyl groups and, especially, the secondary hydroxyl
groups are aligned in the same plane, we thought that a CD
dimer would be a powerful transporter for saccharides if the
CD dimer could encompass the saccharide, with hydrogen
bonds formed between the hydroxyl groups of the saccharide
and the hydroxyl groups of CD at the secondary hydroxyl side,
and if the primary hydroxyl side of CD were chemically
modified with hydrophobic groups. In this paper, we report a
new highly efficient system for transporting saccharides
through a liquid membrane using a CD dimer.
the transport of saccharides through a biomembrane.[1 The
4]
separation of a mixture of saccharides is not easy, because
most saccharides are isomers that differ only in the config-
uration of specific hydroxyl groups. An effective method for
separating a mixture of saccharides is based on transport of
the saccharide through a selective membrane containing
transporters that bind only to specific saccharides. The most
successful transporters bind to saccharide through the selec-
tive formation of hydrogen bonds or covalent bonds. The
transport rate of a saccharide through a hydrophobic liquid
membrane is quite slow, because saccharides are hydrophilic.
However, the transport rate of the saccharide through the
liquid membrane would be increased, if the saccharide could
be covered by a hydrophobic shell that has functional groups
on its inside that interact with the hydroxyl groups of the
saccharide. Phenylboronic acid derivatives, reversed micelles,
and lipophilic alkaline earth metal complexes have been used
Results and Discussion
A new transporter was prepared as shown in Scheme 1.
Saccharides are predicted to be caught effectively by the
second hydroxyl groups of the two b-CDs, which are
connected at the secondary hydroxyl side. All of the primary
hydroxyl groups have to be modified with hydrophobic
groups, because b-CD is insoluble in chloroform but soluble
in water; this solubility is not suitable for a saccharide
transporter. Thus, all of the primary hydroxyl groups of b-CD
were first protected with tert-butyldimethlylsilyl groups in
previously as saccharide transporters.[1 It is reported that a
ternary complex of saccharide, water and resorcinarene is
4]
[a] Dr. H. Ikeda, A. Matsuhisa, Prof. A. Ueno[‡]
Department of Bioengineering
order to alter the primary hydroxyl side of b-CD to a
Graduate School of Bioscience and Biotechnology
Tokyo Institute of Technology, 4259 Nagatsuta-cho
Midori-ku, Yokohama 226-8501 (Japan)
Fax : (‡81)45-924-5833
10]
hydrophobic nature.[9
Because heptakis(6-O-tert-butyldi-
methylsilyl)-b-CD (1) was insoluble in water but was soluble
in chloroform, the silylated b-CD 1 was suitable for use as a
transporter in a chloroform liquid membrane. Two molecules
of silylated b-CD 1 were then linked with bis(2-bromoethyl)
E-mail: hikeda@bio.titech.ac.jp
‡
[ ] Professor Akihiko Ueno passed away on March 23, 2003.
Chem. Eur. J. 2003, 9, 4907 4910
DOI: 10.1002/chem.200304816
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4907

H. Ikeda et al.
FULL PAPER
and methyl a-d-mannopyrano-
side were accelerated more
than fourfold by the addition
of the b-CD dimer 2, but was
accelerated less than twofold by
the addition of the b-CD mon-
omer 1. The transport rate of d-
ribose across the liquid mem-
brane in the presence of the b-
CD dimer 2 was 2.5 times faster
than that in its absence, by
contrast, the transport rate of
d-ribose across the liquid mem-
brane in the presence of the
Scheme 1. a) tBuMe2SiCl, pyridine, 48C, 18 h, 85.5%; b) bis(2-bromoethyl) ether, NaH, THF/DMSO, 508C,
48 h, 10.9%.
ether to obtain the CD dimer 2 with a yield of 10%. After the
reaction mixture was purified by column chromatography, the
product was identified by elemental analysis, different
1H NMR spectra, and mass spectrometry.
The transport experiments were performed in a U-tube
glass cell across a chloroform liquid membrane from an
aqueous source phase containing a saccharide to an aqueous
receiving phase (Figure 1). Three kinds of liquid membranes
Figure 2. Substrates used in the transport experiments.
Figure 1. U-type glass cell used for the transport experiments.
were examined for each saccharide: the chloroform liquid
membrane in the presence of ether 1 or 2, or in their absence.
Three kinds of methyl pyranosides (methyl a-d-glucopyrano-
side, methyl a-d-galactopyranoside, and methyl a-d-manno-
pyranoside), two kinds of pentoses (d-ribose, and d-2-
deoxyribose), and two kinds of disaccharides (d-maltose,
and d-lactose) were used as substrates of the transport
experiments (Figure 2).
Figure 3. Time course of the concentration of methyl a-d-glucopyranoside
in the receiving phase. Source phase (3 mL H2O, [Me-Glu] ˆ 1.0 m); organic
phase (12 mL CHCl3, [1] ˆ [2] ˆ 1.0 mm); receiving phase (3 mL H2O),
258C.
At regular intervals, a small aliquot of the aqueous
receiving phase was sampled, suitably diluted, and assayed
by HPLC. The amount of saccharide transported to the
receiving phase increased linearly with time in the all cases
(Figure 3). The transport rate was determined by the slope of
the time course of the amount of the transported saccharide
(Table 1). The transport rates of methyl a-d-glucopyranoside
b-CD monomer 1 was almost the same as in its absence.
Because d-deoxyribose is more hydrophobic than the other
saccharides and its transport rate in the absence of the
transporter is fast, the difference in transport rates among the
three kinds of membrane condition was small. Nevertheless,
the transport rate of d-deoxyribose was faster in the presence
of 2 than in the presence of 1.
4908
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2003, 9, 4907 4910

Saccharide Transporters
4907 4910
Table 1. Transport rates of saccharides through the liquid membrane
mediated by 1 or 2.
Unfortunately, the b-CD dimer 2 was not effective for
transporting disaccharides. Because disaccharides are them-
selves insoluble in chloroform and it is impossible to transport
disaccharides through the chloroform membrane in the
absence of an effective transporter, it is necessary to wrap
the disaccharide completely by the transporter for passage
through the chloroform layer. If the binding constant of a
transporter for a disaccharide is not large, the disaccharide
will be released in the chloroform layer during the transport
and will subsequently precipitate in this layer. The binding
affinity of the b-CD dimer 2 is sufficient for transporting a
monosaccharide but not a disaccharide. It will be therefore
necessary to introduce effective functional groups to the b-CD
dimer 2 in order to improve the binding ability for disacchar-
ides.
Although the b-CD dimer 2 is not effective for transporting
disaccharides, our experimental results indicate that it func-
tions quite effectively in the liquid membrane as the trans-
porter for the monosaccharides. Whereas the b-CD dimer 2
can almost completely encompass the monosaccharides, the
b-CD monomer 1 is not enough to cover up the monosac-
charides. Although in theory two b-CD monomers could
make a complex with a saccharide in which the saccharide is
covered in a similar manner to that achieved by the b-CD
dimer, the concentration of this complex in the chloroform
layer would be quite small. Therefore, the b-CD dimer is more
effective than the b-CD monomer for the transporting the
saccharide through the liquid membrane. If functional groups
that could interact selectively with the hydroxyl groups of the
saccharide were introduced into the secondary hydroxyl side
of b-CD, then a more efficient transporter might be con-
structed.
Substrate
Transport rate [10À2 mm hÀ1
]
non
1
2
Me-Glu
Me-Gal
Me-Man
Ri
De
Mal
2.9
5.8
3.7
3.4
33.0
0
5.0
6.3
5.5
3.7
36.0
0
12.0
13.7
14.9
8.4
39.0
0
Lac
0
0
0
The transport rate in the absence of transporter was
subtracted from the transport rate in the presence of the
transporter to determine the transport rate of the complex of
the saccharide with the transporter (Figure 4). The transport
rate for the complex of methyl a-d-galactopyranoside with
Figure 4. Transport rates of complexes of monosaccharides with cyclo-
dextrin derivatives through the liquid membrane.
the b-CD dimer 2 was 16 times larger than that with the b-CD
monomer 1. The transport rate for the complex of d-ribose
with the b-CD dimer 2 was 17 times larger than that with the
b-CD monomer 1. The transport rates of the three types of
methyl pyranoside in complex with the b-CD dimer 2 were
almost same. The transport rates of the complex of ribose with
the b-CD dimer 2 are similar to that of deoxyribose with the b-
CD dimer 2. These results can lead to an estimated
mechanism that the saccharides are completely covered by
the b-CD dimer 2 and that the hydrophobicities of these
complexes are almost same, although the structure of the
complex of the saccharide with the b-CD dimer 2 is not
identified.
Because we could not detect a complex of the b-CD dimer
and the saccharide in chloroform by a 1H NMR spectroscopy,
the concentration of this complex in chloroform is not high.
However, the complex is quite effective for transporting
saccharides, even at this low concentration. A successful
transporter has not only to trap a saccharide effectively at the
interface between the aqueous source phase and the chloro-
form liquid membrane, but also to release the saccharide
rapidly at the interface between the aqueous receiving phase
and the chloroform liquid membrane. Therefore, a good
transporter should have a moderate binding affinity for a
saccharide rather than a high binding affinity.
Conclusion
A new and effective system for transporting saccharides
through a liquid membrane has been constructed using a b-
CD dimer as a transporter. This system is more successful for
transporting saccharides through a liquid membrane than is a
system using a b-CD monomer as a transporter.
Experimental Section
General: Mass spectra were obtained on a Shimadzu MALDI-III (TOF-
MS). 1H NMR spectra were recorded on a Varian VXR-500SFT NMR
spectrometer. Chloroform (CHCl3, d ˆ 7.26) was used as an internal
standard, and tetramethylsilane (TMS, d ˆ 0) was used as an external
standard. Elemental analyses were performed by the Analytical Division in
Research Laboratory of Resources Utilization of Tokyo Institute of
Technology. b-Cyclodextrin was a kind gift from Nihon Shokuhin Kako
Ltd. Other reagents were purchased from Tokyo Kasei. CDCl3 with an
isotopic purity of 99.95%, was purchased from Merck.
Heptakis(6-O-tert-butyldimethylsilyl)-b-CD (1): A solution of tert-butyl-
dimethylchlorosilane (14.5 g) in pyridine (150 mL) was added to a solution
of b-CD (9 g) in pyridine (100 mL) for 3.5 h at 08C under nitrogen. The
reaction mixture was stirred for 18 h at room temperature under nitrogen.
After the solvent was removed, the residue was dissolved in chloroform,
washed with 1n hydrochloric acid aqueous solution, saturated aqueous
solution of sodium hydrogen carbonate, and saturated brine, and dried over
magnesium sulfate. The crude product was purified by column chromatog-
Chem. Eur. J. 2003, 9, 4907 4910
www.chemeurj.org
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4909

H. Ikeda et al.
FULL PAPER
raphy on silica gel (chloroform/ethyl acetate 7:0.1) and by recrystallization
from methanol/chloroform (95:5) to yield 13.1 g, 85.5%. 1H NMR
(500 MHz, CDCl3): d ˆ 0.03 (s, 21H; Si-CH3), 0.04 (s, 21H; Si-CH3), 0.87
(s, 63H; C-(CH3)3), 3.56 (t, J ˆ 9.2 Hz, 7H; H-4), 3.61 (m, 7H; H-5), 3.64
(dd, J ˆ 3.3, 9.2 Hz, 7H; H-2), 3.71 (brd, J ˆ 10.0 Hz, 7H; H-6), 3.90 (dd,
J ˆ 3.1, 11.4 Hz, 7H; H-6'), 4.04 (t, J ˆ 9.2 Hz, 7H; H-3), 4.89 (d, J ˆ 3.4 Hz,
7H; H-1); MALDI-TOFMS: m/z: calcd for C84H168O35Si7Na: 1956; found:
Acknowledgements
We are grateful to Nihon Shokuhin Kako Co., Ltd. for a generous supply of
cyclodextrins. This work was supported by a Grant-in-Aid for Science
Research from the Ministry of Education, Culture, Sports, Science and
Technology and by the Grant of the 21st Century COE Program of Ministry
of Education, Culture, Sports, Science and Technology.
‡
1956 [M ‡Na]; elemental analysis calcd (%) for C84H168O35Si7 ¥ H2O ¥
1
³3 CHCl3: C 50.83, H 8.60, Cl 1.78; found: C 50.79, H 8.39, Cl 1.89.
b-CD dimer (2): NaH (18 mg) was added to a solution of 1 (1.0 g) in a
mixed solvent of THF (30 mL) and DMSO (3 mL) and stirred at room
temperature. A solution of bis(2-bromoethyl) ether (57 mg) in THF
(20 mL) was added to the reaction mixture for 10 h at 508C. After the
solvents were removed, the residue was dissolved in chloroform, washed
with saturated aqueous solution of ammonium chloride and dried over
magnesium sulfate. The crude product was purified by column chromatog-
raphy on silica gel (chloroform/methanol/water 5:1:0.1). The fractions
containing the desired product were concentrated under reduced pressure
to give a white powder (221.9 mg, 10.9%). 1H NMR (500 MHz, CDCl3):
d ˆ 0.03 (brs, 42H; Si-CH3), 0.87 (s, 63H; C-(CH3)3), 3.22 (brd, 2H; H-2),
4.82 5.02 (m, 14H, H-1); MALDI-TOFMS: m/z: calcd for C172H342O71-
[1] T. Araki, H. Tsukube, Liquid Membranes: Chemical Applications,
CRC Press, Florida, 1990.
¬
[2] N. Tbeur, T. Rhlalou, M. HlaÔbi, D. Langevin, M. Metayer, J.-F.
¡
Verchere, Carbohydr. Res. 2000, 329, 409.
[3] K. Kasuga, T. Hirose, S. Aiba, T. Takahashi, K. Hiratani, Tetrahedron
Lett. 1998, 39, 9699.
[4] T. Kida, D. Furue, A. Masuyama, Y. Nakatsuji, I. Ikeda, Chem. Lett.
1996, 733.
[5] Y. Aoyama, Y. Tanaka, S. Sugahara,J. Am. Chem. Soc. 1989, 111, 5397.
[6] Comprehensive Supramolecular Chemistry, Vol 3, (Eds.: J. Szejtli, T.
Osa), Pergmon, Oxford, 1996.
[7] J. Szejtli, Cyclodextrin Technology, Kluwer, Dordrecht, 1998.
[8] G. Wenz, Angew. Chem. 1994, 106, 851; Angew. Chem. Int. Ed. Engl.
1994, 33, 803
‡
Si14K: 3965; found: 3965 [M ‡K]; elemental analysis calcd (%) for
C
172H342O71Si14 ¥ 3H2O ¥ 1³3 CHCl3: C 51.32, H 8.70, Cl 0.88; found: C 51.35, H
8.68, Cl 0.90.
Transport experiments: Transport experiments were performed in a U-tube
glass cell across a chloroform liquid membrane from an aqueous source
phase (3 mL) containing a saccharide (1m) to an aqueous receiving phase
(3 mL) (Figure 1). Three kinds of liquid membranes were examined for
each saccharide: chloroform liquid membrane (12 mL) in the presence of
either 1 or 2 (1mm), or in their absence at 258C. At regular intervals, a small
aliquot of the aqueous receiving phase was sampled, suitably diluted, and
assayed by HPLC using Tosoh TSKgel Amide-80. Saccharides were
detected by a refractive index meter.
[9] D. Alker, P. R. Ashton, V. D. Harding, R. Kˆniger, J. F. Stoddart,
A. J. P. White, D. J. Williams, Tetrahedron Lett. 1994, 35, 9091.
[10] M. R. de Jong, J. F. J. Engbersen, J. Huskens, D. N. Reinhoudt, Chem.
Eur. J. 2000, 6, 4034.
Received: February 6, 2003 [F4816]
4910
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2003, 9, 4907 4910