Target-selective degradation of oligosaccharides by a light-activated small-molecule-lectin hybrid

Article abstract of DOI:10.1002/anie.200702127

(Chemical Equation Presented) An innovative method for target-selective degradation of oligosaccharides induced by a light-activated hybrid molecule consisting of anthraquinone and lectin has been developed (see scheme). The reaction takes place under neu

Full text of DOI:10.1002/anie.200702127

Communications
DOI: 10.1002/anie.200702127
Oligosaccharides
Target-Selective Degradation of Oligosaccharides by a Light-Activated
Small-Molecule–Lectin Hybrid**
Maiko Ishii, Shuichi Matsumura, and Kazunobu Toshima*
Oligosaccharides play important roles in many biological
events, including protein folding, cell signalling, fertilization,
pathogen binding to host tissue, leukocyte trafficking and
associated inflammatory response, tumour cell metastasis,
and regulation of hormone and enzyme activity.[1] The
development of innovative methods for selectively control-
ling specific functions of certain oligosaccharides has
attracted much attention in the fields of chemistry, biology,
and medicine. However, there have as yet been no reports of
methods involving specific inhibition of function by selective
degradation of a target oligosaccharide. Herein we report the
target-selective degradation of an oligosaccharide induced by
a light-activated small-molecule–lectin hybrid. To the best of
our knowledge, this is the first successful example of target-
selective degradation of an oligosaccharide by light switching
under neutral conditions.
With the aim of investigating a novel small molecule for
degradation of oligosaccharides, we examined the anthraqui-
none molecule, which is found in several biologically impor-
tant natural products, particularly antibiotics.[2] Certain
anthraquinone derivatives have been found by Schuster and
co-workers and by us to be efficient agents for DNA
photocleavage.[3,4] In addition, photoinduced deterioration
of cellulose by anthraquinone-related dyes has been reported,
although the mechanism and the precise degradation prod-
ucts have not yet been elucidated.[5] On the basis of these
Scheme 1. Anthraquinone derivatives and oligosaccharides. Ac=acetyl.
previous findings, we expected that if an anthraquinone
derivative could be made to produce a radical species by
photoexcitation,[6] this species could be used for degradation
not only of DNA, but also of oligosaccharide molecules.
To investigate this hypothesis, we selected the anthraqui-
none derivative 2-hydroxymethylanthraquinone (1) owing to
its solubility in aqueous media. For the oligosaccharides, we
chose maltoheptaose (2), a-cyclodextrin (a-CD, 3), b-cyclo-
dextrin (b-CD, 4), and g-cyclodextrin (g-CD, 5; Scheme 1).
Although all of these oligosaccharides consist of a-d-glucose,
they may be divided into two groups on the basis of their
affinity for the anthraquinone derivative 1. Thus, the mem-
bers of the first group, maltoheptaose and a-CD, have no
specific interaction with 1, while those in the other group, b-
CD and g-CD, showstrong interaction with 1 owing to their
ability to form an inclusion complex with 1 through hydro-
phobic interaction.[7] Therefore, although a simple anthraqui-
none derivative was used, the oligosaccharides employed
clearly have different affinities for this molecule.
We first examined the photoinduced oligosaccharide-
degrading activity of 1 at a concentration of 300 mm against
30 mm of b-CD (10:1 1/b-CD) in 20% acetonitrile/water
solution, using a long-wavelength UV light source (365 nm,
100 W) for photoirradiation. The progress of the photo-
degradation reaction was monitored by HPLC. It was found
that no change in the HPLC profile was obtained either by
treatment of b-CD with 1 without photoirradiation or by
photoirradiation of b-CD in the absence of 1. In stark
contrast, the HPLC peak corresponding to b-CD completely
disappeared after exposure of b-CD to 1 with photoirradia-
tion, which indicates that degradation of b-CD did take place.
These results clearly showthat this anthraquinone derivative
is capable of degrading an oligosaccharide, b-CD, upon
[*] M. Ishii, Prof. Dr. S. Matsumura, Prof. Dr. K. Toshima
Department of Applied Chemistry
Faculty of Science and Technology
Keio University
3-14-1 Hiyoshi, Kohoku-ku, Yokohama 223-8522 (Japan)
Fax: (+81)45-566-1576
E-mail: toshima@applc.keio.ac.jp
[**] This research was supported in part by the 21st Century COE
Program “Keio Life-Conjugated Chemistry” of the Ministry of
Education, Culture, Sports, Science and Technology of Japan
(MEXT).
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
8396
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 8396 –8399

Angewandte
Chemie
irradiation with long-wavelength UV light and without
further additives. Because no degradation of b-CD by 1
took place in the absence of light, it was confirmed that UV
light functioned as a trigger, initiating oligosaccharide degra-
dation by the anthraquinone derivative. These phenomena
were also confirmed by MALDI-TOF MS analysis. Again, the
MS peak corresponding to b-CD completely disappeared only
after incubation of b-CD with 1 under photoirradiation.
The degradation product(s) could not be specifically
detected by either HPLC or MS analysis at this stage. We
therefore attempted to convert the products into their
acetylated derivatives, which would be more easily detected
and analyzed. After the photoreaction, total acetylation of the
resulting degradation products was carried out with Ac2O and
TFA.[8] Analysis of the acetylated products using MALDI-
TOF MS (Figure 1) clearly indicated the presence of oxida-
tion products including b-CD oxidized at the C-6 position (b-
CD aldehyde),[9,10] and at the C-1 and C-6 positions, the latter
of which results from oxidative cleavage of the glycosidic
bond of b-CD. Thus, in Figure 1, peaks A1–A5 correspond to
the molecular weights of compounds 6–10, respectively, and
peaks B1–B5 correspond to compounds 11–15. Peak C1,
representing compound 16, which results from hydrolysis of
b-CD and subsequent acetylation, was not detected. The
1H NMR spectrum of a mixture of the degradation products
also indicated the presence of an aldehyde function (d =
10.2 ppm, br). Furthermore, it was confirmed that the
degradation products could not be converted back into b-
CD by reduction with NaBH4.[9,10] These results showthat
photoirradiation of b-CD in the presence of the anthraqui-
none derivative 1 caused not only oxidation of the C-6
hydroxy group(s), but also oxidative cleavage of the C-1
glycosidic bond of b-CD.
The oligosaccharide-degrading activity of the anthraqui-
none derivative 1 decreased significantly in the presence of
radical and hydrogen peroxide scavengers EtOH and KI; in
contrast, the singlet oxygen scavenger histidine caused no
inhibition of photodegradation. In addition, it was confirmed
that a hemiacetal, the hydrolysis product of b-CD, was not
produced during photodegradation. Therefore, oligosacchar-
ide degradation (oxidative cleavage of the glycosidic bond)
must arise from a radical species produced by photoexcitation
of anthraquinone and O2, as shown in Scheme 2.[11–13]
Scheme 2. Presumed photodegradation pathway (oxidative cleavage of
glycosidic bond) of oligosaccharide by 1 (shown here as AQ).
Next, we examined the effects of the concentration of 1
and its ability to interact with the oligosaccharide on the
photodegradation reaction. For the former investigation,
photoinduced b-CD (30 mm) degradation was conducted by
using 1 at concentrations of 30, 90, and 300 mm. The progress
of the photodegradation reaction was monitored by HPLC,
and the percentage degradation was calculated on the basis of
the area of the peak corresponding to b-CD. It was found that
the degradation ability of 1 is dependent on its concentration,
and its activity increased as the concentration increased
(Figure 2a). Furthermore, the use of three equivalents of 1 in
the reaction with b-CD was sufficient to cause almost 100%
degradation. Next, we carried out photoinduced oligosac-
charide degradation using 1 (90 mm) with oligosaccharides
other than b-CD: maltoheptaose (2), a-CD (3), and g-CD (5)
at concentrations of 30 mm each. As shown in Figure 2b,
maltoheptaose, a-CD, and g-CD also underwent degradation
by 1 with photoirradiation. These results demonstrate that the
anthraquinone derivative 1 is capable of degrading not only b-
CD, but a range of other oligosaccharides. However, it was
noted that degradation of b- and g-CDs by 1 was much more
effective than that of maltoheptaose and a-CD. These results
clearly showthat the degradation ability of 1 is highly
dependent on its interaction with the oligosaccharides.
Figure 1. MALDI-TOF MS profile of acetylated products obtained by
photodegradation of b-CD and subsequent acetylation. b-CD (30 mm)
was incubated with 1 (100 mm) in 20% acetonitrile/H2O (30 mL) at
258C for 2 h under irradiation with a UV lamp (365 nm, 100 W) placed
10 cm from the mixture, and then acetylated using Ac2O and TFA at
258C for 15 h. The resulting products were analyzed by MALDI-TOF
MS (matrix: 2,5-dihydroxybenzoic acid). The peaks A1–A5 and B1–B5
are observed at m/z 2039.17, 1995.15, 1951.12, 1907.10, 1863.12,
2097.17, 2053.15, 2009.13, 1965.19, and 1921.16, respectively; these
signals correspond to the molecular weights of compounds 6–15.
TFA=trifluoroacetic acid.
Angew. Chem. Int. Ed. 2007, 46, 8396 –8399
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
8397

Communications
ides. For this purpose, we attempted to design a molecule in
which the anthraquinone derivative 1 was attached to a
lectin,[14] which has specific and strong affinity for sugars, to
create a hybrid containing both degradation and recognition
sites for a target oligosaccharide. For this purpose, we chose
peanut agglutinin (PNA)[15,16] as a lectin, and the tumor-
associated disaccharide Gal(b-1,3)GalNAc, generally known
as T-antigen, as the target oligosaccharide.[17] Scheme 1 shows
the designed hybrid molecule 17 (Scheme 3). This molecule,
which possesses an anthraquinone unit at the sugar-binding
pocket of PNA, was synthesized by a modified version of the
procedure reported by Hamachi et al.[18] To carry out binding
of the affinity ligand 24, which was prepared from 21[19] and
23,[20] to the sugar pocket of PNA, a photoreaction was
conducted with UV irradiation. Subsequent treatment of the
labelled PNA 25 with DTT afforded a structure containing a
mercaptobenzyl site (26), which was modified using 2-
bromoacetoxymethylanthraquinone (27) to furnish the
desired hybrid 17.
With our designed anthraquinone–lectin hybrid in hand,
we then examined its application in the target-selective
photodegradation of oligosaccharides. Photoinduced degra-
dation of three types of oligosaccharide (30 mm each)—Glc(b-
1,4)Glc (18), Man(a-1,3)Man (19), and Gal(b-1,3)GalNAc
(20)—was carried out using 17 (9 mm). The progress of the
photodegradation reaction was monitored by HPLC, and the
percentage degradation was calculated based on the peak
area corresponding to each oligosaccharide. The results are
summarized in Figure 2c,d. When 2-hydroxymethylanthra-
quinone (1) was used in the reaction, no degradation was
detected owing to the low concentration of 1 and its low
affinity for the oligosaccharides. However, when the hybrid 17
was exposed to the oligosaccharide Gal(b-1,3)GalNAc (20)
under photoirradiation, significant degradation took place.
This result is in sharp contrast to those of the other
oligosaccharides 18 and 19, which showed no degradation
under photoirradiation with 17. These results clearly indicate
that the anthraquinone–lectin hybrid 17 selectively degraded
only the target tumor-associated disaccharide Gal(b-
1,3)GalNAc upon photoirradiation, without any additives
and under neutral conditions.
Figure 2. Photodegradation of oligosaccharides. a) b-CD (30 mm) was
incubated with 1 (30, 90, and 300 mm) in 20% acetonitrile/H2O
(30 mL) under the conditions described in Figure 1 and analyzed by
HPLC (TSKgel Amide-80, 4.6”250 mm; 3:2 MeCN/H2O; flow rate
1.0 mLminÀ1; 308C; detection by IR). b) Maltoheptaose (2), a-CD (3),
b-CD (4), and g-CD (5) at concentrations of 30 mm each were
incubated with 1 (90 mm) in 20% acetonitrile/H2O (30 mL) under the
conditions described in Figure 2a and analyzed by the method
indicated in Figure 2a. c) Glc(b-1,4)Glc (18), Man(a-1,3)Man (19), and
Gal(b-1,3)GalNAc (20) at concentrations of 30 mm each were incubated
with 1 (9 mm) in phosphate buffer (30 mL, 10 mm, pH 7.4) at 258C for
2 h under irradiation with a UV lamp (365 nm, 100 W) placed 10 cm
from the mixture and analyzed by HPLC (TSKgel Amide-80,
4.6”250 mm; 3:2 MeCN/H2O; flow rate 0.5 mLminÀ1; 308C; detection
by IR). d) Glc(b-1,4)Glc (18), Man(a-1,3)Man (19), and Gal(b-
1,3)GalNAc (20) at concentrations of 30 mm each were incubated with
17 (9 mm) under the conditions described in Figure 2c and analyzed by
the method indicated in Figure 2c.
On the basis of these favorable results, we turned to the
investigation of target-selective degradation of oligosacchar-
Received: May 15, 2007
Revised: July 17, 2007
Published online: September 26, 2007
Keywords: anthraquinone ·
.
cyclodextrins · lectin ·
oligosaccharides · photochemistry
[1] Carbohydrate in Chemistry and
Biology (Eds.: B. Ernst, G. W.
Hart, P. Sinaþ), Wiley-VCH,
Weinheim, 2000.
Scheme 3. Synthesis of anthraquinone–lectin hybrid 17. Reagents and conditions: a) PySSPy, LiOH,
MeOH, 54%; b) AcOH, MeOH, H2O, 84%; c) hn, phosphate buffer (10 mm, pH 7.4); d) DTT, phosphate
buffer (10 mm, pH 8.1, 10 mm EDTA); e) BrCH2COCl, 2,6-lutidine, CH2Cl2, 90%; f) DMF, phosphate
buffer, (10 mm, pH 7.4). Py=pyridyl; DTT=1,4-dithiothreitol.
[2] Dictionary of Antibiotics and
Related Substances (Ed.: B. W.
Bycroft), Chapman and Hall,
London, 1988.
8398
www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 8396 –8399

Angewandte
Chemie
[3] T. Koch, J. D. Ropp, S. G. Sligar, G. B. Schuster, Photochem.
Photobiol. 1993, 58, 554 – 558.
[4] K. Toshima, Y. Maeda, H. Ouchi, A. Asai, S. Matsumura,
Bioorg. Med. Chem. Lett. 2000, 10, 2163 – 2165.
[5] C. Kujirai, T. Shimizu, S. Hashiya, K. Hosono, Kogyo Kagaku
Zasshi 1969, 72, 270 – 275.
[6] J. M. Bruce in The Chemistry of the Quinoid Compounds, Pt. 1
(Ed.: S. Patai), Wiley-VCH, NewYork, 1974, pp. 465 – 538.
[7] Cyclodextrin Chemistry (Eds.: M. L. Bender, M. Komiyama),
Springer, NewYork, 1978.
[8] I. Simkovic, J. Alfoldi, Carbohydr. Res. 1990, 201, 346 – 348.
[9] A. R. Gibson, L. D. Melton, K. N. Slessor, Can. J. Chem. 1974,
52, 3905 – 3912.
[10] A. M. Aquino, C. J. Abelt, K. L. Berger, C. M. Darragh, S. E.
Kelley, M. V. Cossette, J. Am. Chem. Soc. 1990, 112, 5819 – 5824.
[11] B. Armitage, Chem. Rev. 1998, 98, 1171 – 1200.
[12] W. K. Pogozelski, T. D. Tullius, Chem. Rev. 1998, 98, 1089 – 1107.
[13] Free Radicals in Biology and Medicine (Eds.: B. Halliwell,
J. M. C. Gutteridge), Oxford University Press, Oxford, 1985.
[14] H. Lis, N. Sharon, Science 1989, 246, 227 – 234.
[15] R. Lotan, E. Skutelsky, D. Danon, N. Sharon, J. Biol. Chem.
1975, 250, 8518 – 8523.
[16] R. Ravishankar, M. Ravindran, K. Suguna, A. Surolia, M.
Vijayan, Curr. Sci. 1997, 72, 855 – 861.
[17] G. F. Springer, Science 1984, 224, 1198 – 1206.
[18] I. Hamachi, T. Nagase, S. Shinkai, J. Am. Chem. Soc. 2000, 122,
12065 – 12066.
[19] D. Horton, M. J. Miller, Carbohydr. Res. 1965, 1, 335 – 337.
[20] J. F. Reek, S. Bhattacharya, H. G. Khorana, J. Org. Chem. 1993,
58, 7598 – 7601.
Angew. Chem. Int. Ed. 2007, 46, 8396 –8399
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 8399