The effect of structural differences in the reducing terminus of sugars on the binding affinity of carbohydrates and proteins analyzed using photoaffinity labeling

Article abstract of DOI:10.1016/j.bmc.2010.11.067

Because carbohydrates and proteins bind with such low affinity, the nature of their interactions is not clear. Photoaffinity labeling with diazirin groups is useful for elucidating the roles of carbohydrates in these binding processes. However, when carbo

Full text of DOI:10.1016/j.bmc.2010.11.067

Bioorganic & Medicinal Chemistry 19 (2011) 894–899
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
The effect of structural differences in the reducing terminus of sugars
on the binding affinity of carbohydrates and proteins analyzed
using photoaffinity labeling
Isao Ohtsuka a, , Yutaka Sadakane a, Mari Higuchi a, Noriyasu Hada b, Junko Hada b, Nobuko Kakiuchi a,
⇑
Akiyo Sakushima a
a School of Pharmaceutical Sciences, Kyushu University of Health and Welfare, 1714-1 Yoshino-cho, Nobeoka City, Miyazaki 882-8508, Japan
b Faculty of Pharmacy, Keio University, 1-5-30 Shibakoen, Minato-ku, Tokyo 105-8512, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Because carbohydrates and proteins bind with such low affinity, the nature of their interactions is not
clear. Photoaffinity labeling with diazirin groups is useful for elucidating the roles of carbohydrates in
these binding processes. However, when carbohydrate probes are synthesized according to this conven-
tional method, the reducing terminus of the sugar is opened to provide an acyclic structure. Because
greater elucidation of carbohydrate–protein interactions requires a closed-ring carbohydrate in addition
to the photoreactive group, we synthesized new molecular tools. The carbohydrate ligands were synthe-
sized in three steps (glycosylation with allyl alcohol, deprotection, and ozonolysis). Specific binding pro-
teins for carbohydrate ligands were obtained by photoaffinity labeling. Closed ring-type carbohydrate
ligands, in which the reducing sugar is closed, bound to lectins more strongly than open ring-type sugars.
Carbohydrate to protein binding was observed using AFM.
Received 2 November 2010
Revised 28 November 2010
Accepted 30 November 2010
Available online 4 December 2010
Keywords:
Photoaffinity labeling
Molecular tools
Carbohydrate–protein interaction
Atomic force microscopy (AFM)
Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction
Hatanaka synthesized the diazirine units to serve as photoreactive
ligands.5 Because photoreactive units can be easily integrated into
Many biological phenomena are initiated by the recognition of
carbohydrate ligands by endogenous receptor proteins. Glyco-
sphingolipids and glycoproteins in cell membranes are thought
to play especially important roles in a variety of biological events
such as extracellular recognition, cell–cell interaction, differentia-
tion, oncogenesis and immunity.1 Understanding of the role of
carbohydrates could lead to the elucidation of many disease mech-
anisms, but some problems exist in these studies because carbohy-
drates, in contrast with amino acids and nucleotides, are composed
of various complex structures because the configuration of each
hydroxyl group, anomeric configuration to be formed in glycosyla-
tion reactions and the branching, and often bind with low affinity
to proteins.2 If these problems are addressed, elucidation of the
functions of carbohydrates will be accelerated. Formation of a
covalent bond to a photoreactive group allows one to maintain
the complex between the ligand and its binding protein even under
denaturing conditions (Fig. 1).3 Thus, photoreactive groups func-
tion as a powerful hook in fishing for specific binding proteins.
The use of the phenyldiazirine group, one of these photoreactive
groups, seems promising for achieving efficient crosslinking.4
physiological ligands such as peptides,6 proteins,7 DNA,8 and car-
bohydrates9 by a chemoselective reaction, the biochemist can con-
jugate these photoaffinity ligands to their physiological ligand of
interest.10 However, when carbohydrate probes are synthesized
according to this method, the reducing terminus of the sugar is
opened to provide an acyclic structure (1, Fig. 2). This ring opening
has been confirmed by NMR spectral data.9 For glycoconjugates
with relatively small oligosaccharide, both the reducing and non-
reducing end structures are recognized. Oligosaccharides contain-
ing a sugar that has been opened on its reducing end are not rec-
ognized by proteins ( Fig. 3). Therefore, this problem has to be
solved by the development of a new method, and the development
of such a method will promote further elucidation of carbohydrate-
binding protein interactions. In this paper, we report the synthesis
of new molecular tools for the elucidation of carbohydrate function
(2, Fig. 2). These molecular tools contain both closed ring-type car-
bohydrate ligands, in which the reducing end sugar is closed, and
photoreactive diazirine groups for binding to specific proteins.
We examined photoreactive ligand structures as potential photo-
affinity labels for carbohydrate-binding proteins, and we also
investigated the chemical and physical characteristics of carbohy-
drate-protein binding interactions using atomic force microscopy
(AFM).
⇑
Corresponding author. Tel.: +81 982 23 5701; fax: +81 982 23 5702.
E-mail address: ohtsuka@phoenix.ac.jp (I. Ohtsuka).
0968-0896/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2010.11.067

O
O
I. Ohtsuka et al. / Bioorg. Med. Chem. 19 (2011) 894–899
895
N
N
Formation of a covalent bond by
Photoaffinity labeling
F3C
C
Specific recognized
molecular
covalent bond
F3C
R
N
N
F3C
F3C
C
C
C
R1
uv
croslink
N
H
O
~360 nm
R
R
1
Carbohydrate ligand
O
Figure 1. Identification of specific binding molecules for ligands by photoaffinity labeling.
N
N
N
N
F3C
F3C
O
HN NH
HN NH
H
N
H
N
O
O
HO
O
O
O
O
HO
S
S
OH
O
OH
O
O
O
O
OH
OH
HO
OH
O
O
HO
O
N
O
N
HO
HO
OH
OH
OH
OH
1
2
Open Ring Type
Close Ring Type
Figure 2. Target molecular tools containing photoreactive group.
HO
O
O
O
O
OH
O
OH
O
OH
O
OH
Close Ring Type
Open Ring Type
Figure 3. Binding affinity of carbohydrates and proteins are affected by different of the reducing end of sugar.
N-iodosuccimide (NIS) and trifluoromethanesulfonic acid (TfOH)
as the glycosylation promoter.12 After the reaction was com-
pleted as indicated by TLC monitoring, the mixture was purified
by silica gel column chromatography to give pure product 4 in
76% yield. The b-glycosylated linkage was confirmed by 1H
NMR and 13C NMR spectrometry.13 The anomeric proton of the
Glc formed a glycosidic linkage that appeared as a doublet with
a homonuclear coupling constant of 7.8 Hz (d 4.49 ppm, d, 1H,
H-1). Allyl group protons were assigned chemical shifts of
5.90 ppm (m, 1H, –CH@CH2) and 5.11–5.06 ppm (m, 2H, –CH@
CH2). The removal of the acetyl groups in 4 was achieved by its
treatment with 30% NaOMe in MeOH to give 514 in quantitative
2. Results and discussion
2.1. Synthesis of carbohydrate ligand 6
Preparation of the designed carbohydrate ligand 6 was straight-
forward (Scheme 1). We planned a strategy directed to the synthe-
sis of
Disaccharide derivative 4 was obtained by condensation of phe-
nyl-2,3,4,6-tetra-O-acetyl-b- -galactopyranosyl-(1?4)-2,3,6-tri-
O-acetyl-1-thio-b-
-glucopyranoside (3),11 which was prepared
by acetylation and thioglysylation of -lactose. The obtained
a carbohydrate ligand containing an aldehyde linker.
D
D
D
thioglycoside was then treated with allyl alcohol using
1) Ac2O, pyr.
2) PhSH, BF3-Et2O,
CH2Cl2
HO
AcO
RO
Allyalchol, NIS,
TfOH, Ch2Cl2
HO
AcO
AcO
RO
O
O
O
O
O
O
HO
OH
RO
OR
O
SPh
OAc
HO
AcO
RO
O
O
O
OH
OH
OR
OAc
OH
D-lactose
OAc
OR
88% (2steps)
76%
R
3
NaOMe,
MeOH
quant.
Ac
H
4
5
O3, MeOH, then
MeSMe
HO
HO
O
O
HO
OH
O
CHO
HO
O
OH
85%
OH
6
Scheme 1. Syntheis of carbohydrate ligand 6.

896
I. Ohtsuka et al. / Bioorg. Med. Chem. 19 (2011) 894–899
yield. Then selective oxidation to the aldehyde from the alkene in
5 via ozonolysis15 gave a crude mixture containing 6.16 The resi-
due was purified by LH-20 in methanol/water (1:1) to give com-
pound 6 in 85% yield. The structure of carbohydrate ligand 6 was
confirmed by 1H NMR and 13C NMR (d = 79.8 ppm, C-1,
d = 103.5 ppm, C-10) spectroscopy. The aldehyde group of 6 ap-
peared as a singlet at d 8.20 ppm (1H, s) by 1H NMR spectrometry,
and at 181.4 ppm by 13C NMR spectrometry.
proteins, SBA and ECA lectins (Fig. 4). The SBA lectin, which is
obtained from Glycine max (soybean), is known to form the specific
binding for
ECA lectin, which is obtained from Erythrina cristagalli, is also
known to form the specific binding for -lactose, -galactose and
-lactosamine.18 Thus the both labeling tools 1 and 2 should be
D-galactose, D-galactosamine and D
-lactosamine.17 The
D
D
D
bound to both lectins. The labeling tool and lectin protein were
mixed and incubated for 30 min at 37 °C, which resulted in pro-
moting the formation of carbohydrate–protein complex. Then,
the complex was irradiated with 360 nm ultraviolet light to form
the covalent bond between the labeling tool and the lectin. The
complex were separated by SDS–PAGE and blotted on polyvinyli-
dene difluoride membrane. The complex were visualized by
chemiluminescence of luminol reaction which was induced by
horseradish peroxidase (HRP) conjugated with avidin because the
labeling tool had a biotin tag.
The results of photoaffinity labeling experiments with synthetic
labeling tools 1 and 2 were shown in Figure 5. Molecular mass of
complex with SBA and ECA lectin on the membrane were esti-
mated at 32 and 28 kDa, respectively, which were corresponded
to the calculated mass of complexes. The labeling tools 2 recog-
nized more strongly both lectins than tools 1 did because dense
band were observed in lanes 2 and 6 in comparison with lanes 1
and 5, respectively. These results showed that closed ring-type car-
bohydrate labeling tool 2 bind to lectins more strongly than open
ring-type tool 1. In both experiments using SBA and ECA lectins,
addition of excess amount of lactose inhibited the formation of
complex with labeling tool 2 but not that with labeling tool 1
(compare the lanes 1 and 3, 2 and 4, 5 and 7, 6 and 8). These com-
petition experiments showed that only labeling tool 2 recognized
carbohydrate binding site of the lectin. Taken together, the cyclic
form of carbohydrates is important to perform the photoaffinity
labeling with carbohydrate ligands.
2.2. Synthesis of molecular tools 1 and 2
The oxime group was obtained by coupling the aminooxyl group
and aldehyde groups under mild conditions without a promoter. 2-
[2-[2-(Biotynylaminoethoxy)-ethoxy]-ethoxy]-4-[3-(trifluoromet-
hyl)-3H-diazirin-3-yl]-benzyloxyamine (8), which contained the
activated aminooxyl group that, was obtained by the removal of
the t-butoxycarbonyl (Boc) group in affilight-CHO (7, Seikagaku Ko-
gyo, Japan; biotin-N-bocphenylaminodiazirin), was coupled with
compound 6 in acetonitrile/water (3:1), giving the molecular tool
containing both a closed ring-type oligosaccharide structure and a
diazirin group in quantitative yield (2, Scheme 2). Two major peaks
weredetectedbyHPLCanalysisofthereactionresidue, a findingsug-
gested the formation of E–Z oxime isomers. This mixture was ana-
lyzed by 1H NMR, 13C NMR, and HR-FABMS spectrometry. Two
tripletsat d = 7.59 ppm and d = 6.99 ppm were assigned as the oxime
protons of the E- and Z-forms of 2 by 1H NMR spectroscopy.8 Four
doublet peaks were assigned as H-1 (d = 5.13 ppm, 2-Z, d =
5.09 ppm, 2-E) and H-10 (d = 4.27 ppm, 2-E, d = 4.22 ppm, 2-Z). The
E:Z ratio of 2 was also determined to be 1.4:1.0 by 1H NMR. HR-FAB-
MS analysis of compounds 2-E, 2-Z gave a [M+Na]+ ion peak at m/z
993.3310, in agreement with the formula C39H57F3N6O17SNa. The
structures of 2-E, 2-Z were confirmed by these analytical results.
The structural heterogeneity of the oxime isomers of the linker
might be a synthetic drawback. However, the presence of oxime
isomers is likely to be less important for binding photoaffinity la-
bels because the carbohydrate ligands are bound to lectins. In such
a case, the separation of each isomer of the photoprobe may not be
necessary for the application of this method.
2.4. AFM images of carbohydrate and specific binding protein
using photoaffinity labeling
The investigation of the chemical and physical characteristics of
carbohydrate–protein interactions must elucidate the functions of
carbohydrate.19 AFM enables the observation of their surfaces on
the nanoscale level to give insight about their characteristics.
A mixture of molecular tool 2 and SBA lectin was incubated (sam-
ple A) and that irradiated with UV (sample B). Any changes in the
morphology of the surface were observed by AFM. The surfaces of
lectins were flat before UV irradiation (sample A: Fig. 6A), but dot
patterns appeared on them after UV irradiation (sample B: Fig. 6B).
These findings indicated that UV irradiation promoted covalent
We also synthesized compound 1 using D-lactose in a synthesis
similar to that which provided 2.9 The structure of 1-E, 1-Z were
confirmed by 1H NMR, 13C NMR and HR-FABMS.
2.3. Elucidation of carbohydrate–protein interactions by pho-
toaffinity labeling
The binding ability of synthetic labeling tools 1 and 2 were
examined by photoaffinity labeling to two carbohydrate-binding
N N
O
F3C
HN NH
H
N
O
O
O
O
HO
S
CH3CN, H2O
HO
O
6
O
HO
OH
N
O
O
2
HO
O
quant.
OH
OH
N N
O
F3C
R
HN NH
H
N
O
Boc
H
7
8
O
O
O
N N
F3C
O
S
TFA
RHN
O
HN NH
H
N
O
HO
O
O
O
HO
S
OH
OH
CH3CN, H2O
quant.
O
HO
OH
HO
O
N
O
D-lactose
OH
1
Scheme 2. Synthesis of molecular tools 1 and 2.

I. Ohtsuka et al. / Bioorg. Med. Chem. 19 (2011) 894–899
897
Figure 4. Elucidation of molecular tools (1, 2) and specific binding lectin for
D-lactose (SBA, ECA) interactions by photoaffinity labeling.
bond formation between 2 and the lectins. Thus, these dots on the
mountain-like structure imply ligand binding. Because the
molecular size of one SBA lectin is 31 kDa, and amino acids residue
of it is 285 A.A.,17 these images suggested that SBA lectin contains
3. Conclusion
In summary, new molecular tools for the elucidation of carbo-
hydrate roles were developed. These tools featured both a closed
ring-type sugar and a photoactive phenyldiazirine group. We
showed that the structural difference between closed ring- and
open ring-type carbohydrates affected binding affinity to proteins.
many binding sites for
D-lactose. These results provide useful
insight for the elucidation of the physical characteristics of SBA
lectin.

898
I. Ohtsuka et al. / Bioorg. Med. Chem. 19 (2011) 894–899
spectrometry were measured on a LCQ Advantage (Thermo Fisher
97 kD
Scientific Ltd). High-resolution mass spectrometry were measured
on a JMS-700 (JEOL Ltd) under FAB condition. TLC was performed
on silica gel 60 F254 (E. Merk) with detection by quenching UV
fluorescence and by charring with 10% H2SO4. Column chromatog-
raphy was carried out on silica gel 60 (E. Merk). AFM images were
observed using JSPM-5200 (JEOL Ltd).
66 kD
45 kD
31 kD
21 kD
4.2. Chemistry
lane
1
2
3
4
5
6
7
8
4.2.1. 2-Oxyoethyl b-
D-garactopyranosyl-(1?4)-b-D-glucopy-
ranoside (6)
Ally glycoside 5 (0.62 mg 160
l
mol) was dissolved in MeOH
(10 mL). The solution was cooled to ꢀ78 °C and then treated with
ozone and oxygen gas by bubbling until the solution remained
blue. After 10 min, the ozone was evacuated with nitrogen for
15 min at ꢀ78 °C, and dimethyl sulfide (1.0 mL) was added. The
resulting solution was allowed to warm to room temperature over
1 h. The clear solution was concentrated and purified using
Sephadex LH-20 column chromatography (H2O/MeOH = 1:1 as the
eluent) to give 6 (52 mg, 85%). 1H NMR (400 MHz, CDCl3) d 8.20 (s,
1H, –CHO), 4.74 (d, J = 7.8 Hz, 1H, H-1), 4.22 (d, J = 7.8 Hz, 1H,
H-10), 3.69 (t, 1H, H-30), 3.58–3.41 (m, 11H), 3.30 (t, J = 9.6 Hz, 1H,
H-20), 3.20 (t, J = 7.1 Hz, 1H, H-2); 13C NMR (400 MHz, CDCl3) d
181.4 (–CHO), 103.5 (C-10), 79.8 (C-1), 78.4, 77.0, 76.0, 75.7, 73.2,
72.2 (C-2), 71.6 (C-20), 69.3 (C-30), 61.7, 60.6; HR-MS (FAB) calcd
Figure 5. SDS–PAGE analysis of binding affinity of carbohydrates and proteins.
for C14H25Oþ (M+H)+ m/z 385.1346; measure m/z 385.1346.
12
4.2.2. 2-Oxyoethyl b-D-garactopyranosyl-(1?4)-b-D-glucopy-
ranoside-O-2-[2-[2-(biotynylaminoethoxy)-ethoxy]-ethoxy]-4-
[3-(trifluoromethyl)-3H-diazirin-3-yl]-benzyloxime (2-E, 2-Z)
2-[2-[2-(Biotynylaminoethoxy)-ethoxy]-ethoxy]-4-[3-(trifluo-
romethyl)-3H-diazirin-3-yl]-benzyloxycarbanic acid tert—butyl es-
ter (AffiLight—CHO, Seikagaku corporation) (7) (5 mg, 7.3
was dissolved in CH2Cl2 (70 L). To this solution was added triflu-
oroacetic acid (70 L) at 0 °C, and the mixture was stirred for 1 h.
Toluene (100 L) was added to the reaction mixture, and it was
lmol)
l
l
l
then concentrated by nitrogen flash. Triethylamine (18 lL) was
added to the residue and then concentrated to give 8.
To a solution of 8 (4.4 mg, 7.3
lmol) in CH3CN (255
lL) was
added 6 (4.2 mg, 11 mol) in H2O (85
l
lL), and the resulting mix-
ture was shaken for 1 h at room temperature. The mixture of 2-E
and 2-Z (7.1 mg, quant.) was purified by reversed-phase HPLC
using a C18 column, Shiseido Capcell Pack C18 UG120 (20 mm
i.d. ꢁ 250 mm, Shiseido Co. Ltd, Japan, with a linear gradient of
0–80% acetonitrile, 0.1% TFA at a flow rate of 1 mL/min, with mon-
itoring at 215 nm). 1H NMR (400 MHz, CD3OD) d 7.59 (t, 1H,
CH@NO of 2-E), 7.49 (t, 1H), 7.36 (d, J = 7.8 Hz, 1H), 6.99 (br t,
1H, –CH@NO of 2-Z), 6.88 (br d, 1H), 6.74 (br s 1H), 5.13 (s, 1H),
5.09 (s, 1H), 4.47 (d, 1H, –OCH2CH@NO of 2-Z), 4.34 (d, 1H, –
OCH2CH@NO of 2-E), 4.27 (d, 1H, H-10 of 2-E), 4.22 (d, 1H, H-10
of 2-Z), 4.17 (d, 2H, CH@NOCH2– ꢁ 2 of 2-Z), 3.86 (d, 2H,
CH@NOCH2– ꢁ 2 of 2-E); 13C NMR (400 MHz, CDCl3) d 139.7,
119.9, 77.0, 74.8, 71.8, 71.3, 70.6, 70.3, 63.3, 61.6, 56.9, 41.1,
40.3, 36.7, 29.8, 29.5, 26.8, 15.8; HR-MS (FAB) calcd for
Figure 6. AFM Image of compound 2 and SBA lectin: (A) before UV irradiation
(sample A) and (B) after UV irradiation (sample B). Each mount areas are lectin sites.
Scan area: 1.5 ꢁ 1.5
lm.
This study confirmed that carbohydrate structure, in both non-
reducing end and reducing end sugars, is deeply related to the
recognition event between carbohydrates and proteins. These
molecular tools also enable further elucidation of the various car-
bohydrate-binding protein interactions by substituting for the car-
bohydrate ligand in binding studies. As a result, they will be used
in future studies to elucidate carbohydrate functions on cell
surfaces.
C
39H57F3N6O17SNa+ (M+Na)+ m/z 993.3351; measure m/z 993.3310.
4.2.3. 2-Oxyoethyl b- -garactopyranosyl-(1?4)-b- -glucopy-
D
D
ranoside-O-2-[2-[2-(biotynylaminoethoxy)-ethoxy]-ethoxy]-4-
[3-(trifluoromethyl)-3H-diazirin-3-yl]-benzyloxime (1-E, 1-Z)
4. Experimental
To a solution of 7 (4.4 mg, 7.3
lmol) in CH3CN (255
lL) was
4.1. General method
added -lactose (5.4 mg, 11 mol) in H2O (85 l
D
l
L), and the resulting
mixture was shaken for 1 h at room temperature. The mixture of
1-E and 1-Z (9 mg, quant.) was purified by reversed-phase HPLC
using a C18 column, Shiseido Capcell Pack C18 UG120 (20 mm
i.d. ꢁ 250 mm, Shiseido Co. Ltd, Japan, with a liner gradient of
1H NMR and 13C NMR spectra were measured with a JMN AL
400 FT NMR spectrometer (JEOL Ltd) with Me4Si as the internal
standard for solutions in CDCl3, D2O and CD3OD. ESI mass

I. Ohtsuka et al. / Bioorg. Med. Chem. 19 (2011) 894–899
899
0–80% acetonitrile, 0.1% TFA at a flow rate of 1 mL/min, with mon-
itoring at 215 nm). 1H NMR (400 MHz, CD3OD) d 7.47 (t, 1H,
CH@NO of 1-E), 7.38 (t, 1H, –CH@NO of 1-Z), 6.88 (br d, 1H),
6.74 (br s 1H), 5.14 (s, 1H), 5.08 (s, 1H), 4.36 (d, 1H, J = 7.8 Hz, H-
10 of 1-Z), 4.29 (d, 1H, J = 7.8 Hz, H-10 of 1-E); 13C NMR (400 MHz,
CDCl3) d 130.6, 130.3, 119.9, 110.5, 104.9, 77.0, 74.7, 72.5, 71.8,
71.2, 70.6, 70.5, 70.2, 69.5, 63.3, 61.6, 56.9, 49.6, 40.3, 36.7, 29.7,
29.5, 26.8; HR-MS (FAB) calcd for C37H55F3N6O16SNa+ (M+Na)+ m/
z 951.3245; measure m/z 951.3203.
Ministry of Education, Culture, Sports, Science and Technology of
Japan to Y.S.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2010.11.067.
References and notes
4.3. Analysis of binding affinity of carbohydrates and lectins by
photoaffinity labeling
1. Taylor, M. E.; Drickamer, K. Introduction to Glycobiology; Oxford Univ. Press:
New York, 2003.
The lectins (2.5 lg) were obtained from Soybean, G. max and
2. (a) Lee, Y. C. Ciba Found. Symp. 1989, 145, 80–95; (b) Weis, W. I.; Drickamer, K.
Ann. Rev. Biochem. 1996, 65, 441–473; (c) Rini, J. M. Annu. Rev. Biophys. Biomol.
Struct. 1995, 24, 551–557.
3. Hatanaka, Y.; Sadakane, Y. Curr. Top. Med. Chem. 2002, 2, 271–288.
4. Hatanaka, Y.; Nakashima, H.; Kanaoka, Y. Rev. Heteroat. Chem. 1996, 14, 213–
243.
5. Hatanaka, Y. J. Synth. Org. Chem. Jpn. 1998, 56, 53–62.
6. Nakashima, H.; Hashimoto, M.; Sadakane, Y.; Tomohiro, T.; Hatanaka, Y. J. Am.
Chem. Soc. 2006, 128, 15092–15093.
7. Sadakane, Y.; Takagi, T.; Tomohiro, T.; Tsurusawa, K.; Konoha, K.; Kawahara,
M.; Hatanaka, Y. Photomed. Photobiol. 2004, 35–40.
E. cristagalli (J-Oil Mills Co. Ltd, Japan) and incubated with photore-
active carbohydrates 1 and 2 in 50 mM phosphate buffer (pH 7.4)
at37 °Cfor 10 mininthedark. Inthecase ofcompetitorexperiments,
50 mM lactose was pre-incubated with the solution of lectins for
15 min at 37 °C before the addition of photoreactive carbohydrates.
Specimens were then irradiated at 0 °C for 30 min with a 30-watt
long-wavelength UV lamp to promote crosslinking. The irradiated
samples were subjected to SDS–PAGE (12% polyacrylamide gel)
and transferred to a PVDF membrane. Biotinylated carbohydrates
on the membrane were then visualized by chemiluminescent
detection with avidin-horseradish peroxidase (HRP) conjugate.
8. O’Neil, K. T.; Wolfe, Jr. H. R.; Erickson-Viita-nen, S.; Degra, W. F. Science 1987,
236, 1454–1456.
9. Hatanaka, Y.; Kempin, U.; Jong-Jip, P. J. Org. Chem. 2000, 65, 5639–5643.
10. (a) Koshi, Y.; Nakata, E.; Hamachi, I. Org. Biomol. Chem. 2005, 6, 1349–
1352; (b) Fleming, S. A. Tetrahedron 1995, 51, 12479–12520; (c) Fang, K.;
Hashimoto, M.; Jockusch, S.; Turro, N. J.; Nakanishi, K. J. Am. Chem. Soc.
1998, 120, 8543–8544.
11. Sun, B.; Pukin, A. V.; Visser, G. M.; Zuilhof, H. Tetrahedron Lett. 2006, 47,
7371–7374.
12. Veeneman, G. H.; van Boom, J. H. Tetrahedron Lett. 1990, 31, 1331–1334.
13. Matsuoka, K.; Terabatake, M.; Umino, A.; Esumi, Y.; Hatano, K.; Terunuma, D.;
Kazuhara, H. Biomolecules 2006, 7, 2274–2283.
14. Okamoto, K.; Kondo, T.; Goto, T. Tetrahedron 1987, 43, 5919–5928.
15. Sun, X.-L.; Kanie, Y.; Guo, C.-T.; Kanie, O.; Suzuki, Y.; Wong, C.-H. Eur. J. Org.
Chem. 2000, 2643–2653.
16. Kononov, L. O.; Cheshev, P. E.; Nifant’ev, N. E. Russ. Chem. Bull. 2000, 7,
1305–1309.
17. (a) Lis, H.; Sharon, N. J. Biol. Chem. 1978, 253, 3468–3476; (b) Vodkin, L.
O.; Rhodes, P. R.; Goldberg, R. B. Cell 1983, 34, 1023–1031; (c) Buts, L.;
Dao-Thi, M. H.; Loris, R.; Wyns, L.; Etzler, M.; Hamelryck, T. J. Mol. Biol.
2001, 309, 193–201; (d) Olsen, L. R.; Dessen, A.; Gupta, D.; Sabesan, S.;
Sacchettini, J. C.; Brewer, C. F. Biochemistry 1997, 36, 15073–15080; (e)
Dessen, A.; Gupta, D.; Sabesan, S.; Brewer, C. F.; Sacchettini, J. C.
Biochemistry 1995, 34, 4933–4942; (f) Galbraith, W.; Goldstein, I. J. FEBS
Lett. 1970, 9, 197–201; (g) Jaffe, C. L.; Ehrlich-Rogozinski, S.; Sharon, N.
FEBS Lett. 1977, 82, 191–196.
4.4. AFM observation
Sample A—1.0 mM compound 2 in water (2.5
lectin in water (2.5 L) and 0.5 M phosphate buffer (pH 7.6)
(5.0 L) were dissolved in water (30 L).
Sample B—1.0 mM compound 1 in water (2.5
lectin in water (2.5 L) and 0.5 M phosphate buffer (pH 7.6)
(5.0 L) were dissolved in water (30 L).
Solutions of sample A and B (0.1 L) were added dropwise to a
lL), 1.0 mM SBA
l
l
l
lL), 1.0 mM SBA
l
l
l
l
freshly cleaved mica plate (5.0 ꢁ 5.0 mm) and dried for observa-
tion by AFM. The AFM probe used was a Micro Cantilever CSC38
(JEOL Ltd) made of silicon and coated with Au, which had a spring
constant of 0.08 N/m, a length of 250
lm, and a thickness of
1.0 m. The AFM observation was carried out with the contact
l
mode in the air at 25 °C.
18. (a) Iglesias, J. L.; Lis, H.; Sharon, N. Eur. J. Biochem. 1982, 123, 247–252; (b)
Itakura, Y.; Nakamura-Tsuruta, S.; Kominami, J.; Sharon, N.; Kasai, K.;
Hirabayashi, J. J. Biochem. 2007, 142, 459–469; (c) Svensson, C.; Teneberg, S.;
Nilsson, C. L.; Kjellberg, A.; Schwarz, F. P.; Sharon, N.; Krengel, U. J. Mol. Biol.
2002, 321, 69–83.
Acknowledgements
This study was supported by Mitsubishi Chemical Corporation
Fund to I.O., and by Grant-in-Aid for Scientific Research from the
19. Ohtsuka, I.; Yokoyama, S. Chem. Pharm. Bull. 2005, 53, 42–47.