Electrophoretically mediated microscale reaction of glycosidases: Kinetic analysis of some glycosidases at the nanoliter scale

Article abstract of DOI:10.1016/S0008-6215(02)00135-0

Capillary electrophoresis (CE) is one of the extremely important analytical techniques known for its high sensitivity and resolution. We have investigated electrophoretically mediated microanalysis (EMMA) for the assay of some native glycosidases. Under optimized conditions, the enzymatic reactions of α-glucosidase, β-galactosidase and β-N-acetylglucosaminidase were carried out, and the Michaelis constants were obtained. The current method may have advantages over traditional assay methods, especially in terms of the amount of enzyme and substrate required for a reaction.

Full text of DOI:10.1016/S0008-6215(02)00135-0

Carbohydrate Research 337 (2002) 1757–1762
www.elsevier.com/locate/carres
Electrophoretically mediated microscale reaction of glycosidases:
ꢀ
kinetic analysis of some glycosidases at the nanoliter scale
Yoshimi Kanie, Osamu Kanie*
Mitsubishi Kagaku Institute of Life Sciences (MLTILS), Machida-shi, Tokyo 194-8511, Japan
Received 11 February 2002; accepted 5 June 2002
Abstract
Capillary electrophoresis (CE) is one of the extremely important analytical techniques known for its high sensitivity and
resolution. We have investigated electrophoretically mediated microanalysis (EMMA) for the assay of some native glycosidases.
Under optimized conditions, the enzymatic reactions of a-glucosidase, b-galactosidase and b-N-acetylglucosaminidase were
carried out, and the Michaelis constants were obtained. The current method may have advantages over traditional assay methods,
especially in terms of the amount of enzyme and substrate required for a reaction. © 2002 Elsevier Science Ltd. All rights
reserved.
Keywords: Microreaction; Electrophoretically mediated microanalysis; PNP, glycosides; Glycosidases, kinetic analysis
1
. Introduction
tion. When two compounds forming individual plugs
have different electrolytic mobility, the reaction pro-
ceeds while one of the components is passing the other.
The reaction product is then resolved based on the
individual electromobilities in the electric field and
passes through the detector on the way to the end of
the capillary.
Capillary electrophoresis is extremely useful because
of its high sensitivity and resolution, and its importance
as a microanalytical method has been demonstrated.
This largely relies on recent advances in automated
systems that make reproducible injections possible. An
approach in which the capillary is used not only as
isolation field, but also as a field for the reaction is
Oligosaccharides exist on the cell surface and in the
extracellular matrix as a part of glycoproteins and
glycolipids, and they play important roles in biological
1
–17
recognized to be highly valuable.
trophoretically mediated microanalysis (EMMA)
In particular, elec-
6
–17
18
is
phenomena. To investigate the functions of oligosac-
charides, methodological investigations, especially ones
that are applicable to the microscale experiment is of
extreme importance. We have been interested in carbo-
hydrate-related enzymes, especially oligosaccharide-
processing enzymes, and reported some utilities of
advantageous because the amounts of reactants, en-
zyme and substrate, which are often extremely valuable
and expensive, can be minimized.
In general, there are two ways for mixing two com-
ponents in a capillary under electrophoretic conditions.
One is the continuous format,
capillary filled with one of the reactants, and then
another reactant is injected. The other is plug–plug
6
–10,12,13
which uses the
capillary zone electrophoresis (CZE) for the analysis of
19–23
these enzymatic reactions.
In our previous papers,
we have analyzed the reaction kinetics of glycosyltrans-
ferase and glycosidase based on isolation of the sub-
strate and the product, which was possible using borate
buffer to introduce negatively charged functionality
based on the formation of borate ester of hydroxyl
1
1,12,14–17
format,
which relies on a plug–plug interac-
ꢀ
Taken, in part, from the thesis research of Y.K. at Gifu
University, 2002.
*
Corresponding author. Tel.: +81-42-7246238; fax: +81-
24
groups. Through these studies, we realized the neces-
sity to carry out enzymatic reactions in a capillary in
order to minimize consumption of the reaction mixture
43-7246317
E-mail address: kokanee@libra.is.m-kagaku.co.jp (O.
Kanie).
0
008-6215/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0008-6215(02)00135-0

1
758
Y. Kanie, O. Kanie / Carbohydrate Research 337 (2002) 1757–1762
because the reaction volume could not be reduced less
than ꢀ20 mL in our hands due to the experimental
errors associated with surface tension problems when
handling a tiny amount of liquid.
We have addressed this issue using the plug–plug
format under electrophoretically mediated microscale
reaction conditions. a-Glucosidase (EC 3.2.1.20), b-
galactosidase (EC 3.2.1.23) and b-N-acetylglu-
cosaminidase (EC 3.2.1.30) catalyzed reactions were
selected to show eligibility of the method. The analyti-
cal processes were carried out in an automated manner
in order to speed-up the laborious handling.
according to the equipment’s specification. The viscos-
ity (p) was measured using a viscometer (Ubbelohde)
and shown to be p=0.00756 and 0.00810 (Poise) for
the enzyme plug and the substrate plug, respectively.
Procedures for EMMA
Obtaining kinetic parameter of h-Glc-ase. A solution
containing a-Glc-ase (0.05 mg/mL) in 40 mM phos-
phate buffer (sodium phosphate, pH 7.0) was intro-
duced by pressure injection for 3 s into the capillary
equilibrated with 40 mM borate buffer. Then a solution
of PNP-a-Glc at various concentrations in 200 mM
phosphate buffer (pH 7.0) containing uridine (0.2 mM)
as an internal standard was introduced by means of
pressure injection for 5 s, followed by introduction of
200 mM phosphate buffer (pH 7.0) for 1 s. The assay
was carried out by applying an electric field at a
controlled temperature of 37 °C. EMMA was carried
out under conditions where the reaction conversion was
exceeding 10%.
Obtaining the kinetic parameter of i-Gal-ase. Condi-
tions for EMMA were identical except for the following
parameters. b-Gal-ase (0.5 mg/mL) was used. The pH
of phosphate buffers to dissolve enzyme and substrate
were 4.5 and 4.2, respectively.
Obtaining the kinetic parameter of i3-G1cNAc-ase.
Conditions for EMMA were identical except for the
following parameters. The enzyme, b-GlcNAc-ase (0.4
mg/mL) was dissolved in 40 mM phosphate buffer (pH
4.5), and the substrate was dissolved in 80 mM phos-
phate buffer (pH 4.4).
Automation of analytical process.—The automation
cycle consists of (1) washing with 0.1 N NaOH (2 min),
[and regenerated with water (1 min)]; (2) equilibration
of electrolyte buffer (2 min); (3) injection of a solution
containing enzyme in 40 mM phosphate buffer (sodium
phosphate) (3 s); (4) introduction of 200 mM or 80 mM
phosphate buffer for 1 s and; (5) injection of a solution
of substrate at a various concentration in 200 mM or
80 mM phosphate buffer containing uridine (0.2 mM)
(5 s). Each electropherogram was recorded over 4 min.
One cycle took about 12 min including recalibration of
PDA. The stock solutions of enzyme, substrate and
buffers used in the EMMA were kept at 20 °C at the
autosampler reservoir because large difference in tem-
perature between stocks and the capillary (when stored
enzyme at 4 °C) affected the enzymatic reaction. At this
temperature, enzyme activity remained unaffected dur-
ing ca. 5 h required for collecting a series of the
EMMA data.
2. Experimental
Materials.—The source of enzymes and substrates
are as follows. a-Glucosidase (a-Glc-ase; EC 3.2.1.20)
from Saccharomyces sp.: Toyobo Co., Ltd. (Osaka,
Japan), b-galactosidase (b-Gal-ase; EC 3.2.1.23) from
Aspergillus oriza and b-N-acetylglucosaminidase (b-
GlcNAc-ase; EC 3.2.1.30) from bovine kidney: Sigma
Chemical Co. (St. Louis, MO, USA); p-nitrophenyl-a-
D
-glucopyranoside (PNP-a-Glc): Nacalai Tesque., Inc.
Kyoto, Japan); p-nitrophenyl-b- -galactopyranoside
PNP-b-Gal) and p-nitrophenyl-b- -N-acetylglu-
(
(
D
D
cosaminide (PNP-b-GlcNAc): Sigma Chemical Co. (St.
Louis, MO, USA); uridine: Kohjin Co. Ltd. (Tokyo,
Japan). Double-deionized water was prepared from a
Milli-Q system from Millipore Corp. (Milford, MA,
USA).
Instrumentation.—EMMA assays. EMMA assays
were performed on a Beckman P/ACE System 5500
(
USA), which was equipped with a 75 mm i.d. (r)
uncoated fused silica capillary with 37 cm total length
L). Detection was carried out by on-column measure-
(
ment of UV absorbance at 214 and 405 nm with
photodiode array (PDA) at 7.0 cm from the cathode.
Electrophoresis was performed at 18 kV using a 40 mM
borate buffer (pH 9.2) as an electrolyte.
Photometric assays. Assay solutions were analyzed by
a Waters Quanta 4000E capillary electrophoresis sys-
tem, which was equipped with a 75 mm i.d. (r) uncoated
fused silica capillary with 37 cm–L. Detection was
carried out by on-column measurement of UV absorp-
tion at 405 nm at 7.5 cm from the cathode. Samples
were loaded by means of hydrostatic pressure at 10 cm
height for 30 s (ca. 38.4 nL). Electrophoresis was
performed at 15 kV using 40 mM sodium borate (pH
9
3
.4) as an electrolyte at a constant temperature of
7 °C. Electropherograms were recorded on a Millen-
Photometric assay by CE.—Kinetic analysis of h-Glc-
ase. In order to confirm reliability of data obtained by
the EMMA method, a traditional photometric assay
was carried out. Incubations were performed in a total
volume of 20 mL. Unless otherwise stated, reaction
mixtures contained 40 mM phosphate buffer (pH 7.0),
various amount of PNP-a-Glc (0.1–1.0 mM)
nium 2010 system from Millipore Corp. The capillary
used was pretreated or regenerated with 0.1 M NaOH
(
2 min) and elution with buffer before each injection.
Parameters.—The parameters used to estimate vol-
−
2
ume (V) of each plug are: DP=34,475 (dyne cm
)

Y. Kanie, O. Kanie / Carbohydrate Research 337 (2002) 1757–1762
1759
with 5 mU of a-Glc-ase. The reaction was started by
terminated by addition of 40 mL of 0.2 M Na CO .
2 3
the addition of a-Glc-ase, and the reaction mixture was
Kinetic analysis of i-Gal-ase. The procedure is the
same as that described for the analysis of a-Glc-ase
except for the pH of reaction mixtures (pH 4.5), except
for one pH of reaction mixtures (pH 4.5), the substrate
PNP-b-Gal (0.2–2.0 mM), and the enzyme b-Gal-ase (4
mU).
incubated for 5 min at rt. Then, the reaction was
Kinetic analysis of i-GlcNAc-ase. The procedure is
same as that described for the assay of a-Glc-ase,
except for the pH of reaction mixtures pH 4.5), the
substrate PNP-b-GlcNAc (0.5–3.0 mM), and the en-
zyme b-GlcNAc-ase (1.3 mU).
3
. Results and discussion
General aspects.—Among many important carbohy-
drate processing enzymes, several glycosidases namely,
a-G1c-ase, b-Gal-ase and b-GlcNAc-ase in the current
study to demonstrate the feasibility of our method. The
series of enzymes hydrolyze the corresponding p-nitro-
Fig. 1. Schematic representation of a microreaction in a
capillary under electrophoretic conditions. The isolated plugs
of A and B are introduced into capillary. The order of the
injection is determined based on the relative electromobility,
phenyl
D-glycosides as the substrates to yield p-nitro-
phenol (PNP-ol).
m
and m , as shown by the arrows. The reaction proceeds
A
B
Several issues have to be addressed for this EMMA
technique. In order for the enzymatic reaction to take
place in EMMA using the plug–plug format, one of the
reactants must pass the other (Fig. 1). The fast-moving
plug (A) electrophoretically migrates to the end
boundary of slow-moving plug (B) where interaction of
two substances such as an enzyme and its substrate
starts. The reaction finishes when the end boundary of
the plug A reaches front boundary of the plug B.
Furthermore, it is important to select a certain plug
format to enable both successful enzymatic reaction
and isolation of essentially non-charged compounds
such as carbohydrates. For this, we decided to use a
combination of a phosphate buffer with pH close to the
optimal pH of the hydrolyases for the reactions, and
basic borate buffer pH 9.2, as it is commonly used for
the analysis of carbohydrates using capillary zone elec-
trophoresis (CZE) for the purpose of isolation of the
reactants and product, as well as the detection of the
PNP-ol.
Determining the order of injection.—The order of
injection was first determined. Each enzyme migrated
slower than its substrate in borate buffer (pH 9.2) (data
not shown). This suggested that the analysis could be
performed under these conditions where the slow-mov-
ing enzyme is injected first and the fast-moving sub-
strate next injected to enable plug–plug interaction to
occur. It was anticipated that accuracy and reproduci-
bility would be an issue in this microscale reaction;
therefore, as an internal standard, uridine, was used.
Typical electropherograms of individual enzyme reac-
tions are shown in Fig. 2 where the substrate, uridine as
an internal standard and PNP-ol are clearly resolved.
during plug–plug mixing, and the individual plugs of sub-
strate, enzyme and products are separated.
Fig. 2. Electropherograms of enzymatic reactions of glycosi-
dases. (A) Injection of 1.0 mM PNP-a-Glc in 200 mM
phosphate buffer (pH 7.0) after injection of 0.05 mg/mL
a-Glc-ase in 40 mM phosphate buffer (pH 7.0) (3 s) and 200
mM phosphate buffer (pH 7.0) (1 s). (B) Injection of 1.0 mM
PNP-b-Gal in 200 mM phosphate buffer (pH 4.2) after injec-
tion of 0.1 mg/mL b-Gal-ase in 40 mM phosphate buffer (pH
4.5) (3 s) and 200 mM phosphate buffer (pH 4.2) (1 s). (C)
Injection of 1.0 mM PNP-b-GlcNAc in 80 mM phosphate
buffer (pH 4.4) after injection of 0.05 mg/mL b-GlcNAc-ase
in 40 mM phosphate buffer (pH 4.2) (3 s) and 80 mM
phosphate buffer (pH 4.4) (1 s). ꢁ, substrate; ꢂ, product
(
PNP-ol); ", internal standard (uridine). Electrophoresis con-
ditions were as described in the Experimental section (ab-
sorbance: 214 nm). Each peak in the electropherogram was
identified by analyzing individual material prior to the exper-
iments.

1
760
Y. Kanie, O. Kanie / Carbohydrate Research 337 (2002) 1757–1762
triangle) were not affected significantly by the concen-
tration of phosphate buffer as compared with the rela-
tive area obtained for the PNP-ol (open circle; 0.05 mM
and closed circle; 0.005 mM). This indicates that the
enzymatic reaction is independent of the phosphate
concentrations. In the case of b-G1cNAc-ase (open
square), however, the curve did not match the ones
obtained for PNP-ol, which indicates the enzyme reac-
tion was affected by the phosphate concentration and
was efficient at lower concentrations. To obtain optimal
results, we chose a concentration of 200 mM phosphate
buffer for a-Glc-ase and b-Gal-ase and 80 mM buffer
for b-GlcNAc-ase. The use of higher phosphate concen-
trations (\300 mM) were basically avoided due to the
Fig. 3. The effect of concentration of the phosphate buffer on
relative area of PNP-ol against uridine. Result of a-Glc-ase
25
inaccuracy associated with a baseline disturbance.
(
7
ꢃ). a-Glc-ase (0.1 mg/mL) in 40 mM phosphate buffer (pH
Changes in the referenced PNP-ol area at different
buffer concentrations cannot be explained yet, but it
might relate to the peak shape which occurs as the
result of the concentration of substance.
.2) was injected (3 s), followed by injection of various
concentrations of phosphate buffer (pH 7.2) (1 s) and 0.5 mM
PNP-a-Glc in various concentrations of phosphate buffer (pH
7
4
.2) (5 s). Results of b-Gal-ase: (ꢄ) b-Gal-ase (0.1 mg/mL) in
0 mM phosphate buffer (pH 4.5) was injected (3 s), followed
Obtaining kinetic parameters.—Based on the condi-
tions for EMMA of a-Glc-ase described above, b-Gal-
ase and b-GlcNAc-ase catalyzed enzymatic reactions,
we worked further to obtain Km values of these reac-
tions. The reaction time is not directly given due to the
heterogeneous nature of the reaction during the plug–
plug interaction, but a linear relationship was obtained
for the product area during the experiments at a fixed
concentration of phosphate buffer. Thus, we used the
area as an equivalent value of 6 to obtain the constant.
In the case of the a-Glc-ase catalyzed reaction, the
capillary was filled with 40 mM borate buffer (pH 9.2)
and sequentially injected with an enzyme plug in 40
mM phosphate buffer, 200 mM phosphate buffer (pH
7.2), and a plug containing various amounts of sub-
strate in 200 mM phosphate buffer (pH 7.0), all by
means of positive pressure at the anodic end. The
capillary was then placed in the 40 mM borate buffer.
The assay was performed in quadruplicate using a
37-cm capillary with a controlled capillary temperature
at 37 °C and detection at 6.5 cm from the cathodic end.
The reaction was monitored at 214 and 405 nm to
eliminate ambiguity due to the possible effects on the
PNP-ol peak by other factors.
by injection of various concentrations of phosphate buffer
pH 4.2) (1 s) and 1.0 mM PNP-b-Gal in various concentra-
tions of phosphate buffer (pH 4.2) (5 s). Results of b-Glc-
NAc-ase: (ꢅ) b-GlcNAc-ase (0.1 mg/mL) in 40 mM
phosphate buffer (pH 4.5) was injected (3 s), followed by
injection of various concentrations of phosphate buffer (pH
(
4
.2) (1 s) and 1.0 mM PNP-b-GlcNAc in various concentra-
tions of phosphate buffer (pH 4.2) (5 s). 0.05 mM PNP-ol
(
ꢂ), 0.005 mM PNP-ol (ꢁ) (absorbance: 214 nm).
Optimization of EMMA conditions.—We examined
the adsorption of the analytes to the capillary wall by
using polyacrylamide coated and uncoated capillaries,
but we could not find any marked difference in product
2
5
formation during EMMA. We, therefore, decided to
use an uncoated capillary in this study for convenience.
The volume of the enzyme and the substrate involved
in each reaction was estimated based on the equation of
2
6
Poiseuille (Eq. (1)) ,
4
V=DPyr t/8pL
(1)
where DP is the pressure drop across the capillary
during injection, r is a capillary radius, t is injection
time, p is a viscosity of the buffer and L is total length
of the capillary from inlet to outlet. The enzyme and
substrate solutions were injected by pressure for 3 and
The K values were obtained by a Lineweaver–Burk
m
plot, where relative area was used to indicate product
concentrations, which was used instead of 6 values (Fig.
4). Approximately a threefold improvement in sensitiv-
ity was observed for the data obtained at 405 nm. The
5
7
s, respectively. The internal diameter of capillary was
5 mm. EMMA was performed at 37 °C. The length
and volume (V) of the enzyme and substrate plugs are
estimated to be 6.5 and 10 mm and 29 and 45 nL,
respectively, which were isolated by a 2 mm (9 nL)
phosphate buffer plug to avoid cross contamination
during injections.
Km values thus obtained were K =0.9 mM for a-Glc-
m
ase (n=4, each concentration, CV (214 nm)=2.8–
8.2%), K =1.0 mM for b-Gal-ase (n=4, each
m
concentration, CV (214 nm)=2.0–7.9%) and K =1.1
m
It was important to use the proper concentration of
phosphate buffer for the successful enzymatic reac-
tion. As shown in Fig. 3, the transformations in the
mM for b-GlcNAc-ase (n=4, each concentration, CV
(214 nm)=4.3–12.6%), which were in the same range
as the values obtained by the photometric assay shown
below. The results were also similar to the reported
2
5
case of a-Glc-ase (closed square) and b-Gal-ase (open

Y. Kanie, O. Kanie / Carbohydrate Research 337 (2002) 1757–1762
1761
Fig. 4. Double-reciprocal plots of glycosidase assay. (A) a-Glc-ase assay results of the microreaction. (B) b-Gal-ase assay results
of the reaction. (C) b-GlcNAc-ase assay results of the reaction. Concentration of the product [P] (relative area) was used instead
of initial velocity (see text for explanation). Data are obtained in quadruplicate as shown in the figure. Curves were obtained using
means (") of individual data sets.
2
0,22
value.
K values obtained by the photometric assay
galactosidase and b-N-acetylglucosaminidase were suc-
cessfully analyzed, and Km values similar to those
obtained by photometric assay were obtained. This
indicates that EMMA with plug–plug format can be
used as a reliable method for the kinetic analysis of
carbohydrate-related enzymes. Further, the process was
carried out automatically using a temperature con-
trolled autosampler in order to eliminate routine han-
dling and to speed up the process.
m
method are: a-Glc-ase: K =0.790.1 mM (n=3, each
m
concentration, CV=1.5–9.4%), Vmax=1.5 mM/s; b-
Gal-ase: K =1.290.2 mM (n=3, each concentration,
m
CV=1.7–4.9%), Vmax=2.5 mM/s; and b-GalNAc-ase:
K =1.890.3 mM (n=3, each concentration, CV=
.2–10.0%), Vmax=1.6 mM/s, respectively. The Vmax
values were not obtained because it was impossible at
this time to estimate reaction time due to the hetero-
geneity of the buffer system. Furthermore, the mobility
of the entire electrolyte in the capillary would have
been affected.
m
3
References
1
2
3
. Heegaard, N. H. H.; Kennedy, R. T. Electrophoresis
999, 20, 3122–3133.
. Heegaard, N. H. H.; Nilsson, S.; Guzman, N. A. J.
Chromatogr. B 1998, 715, 29–54.
. Colton, I. J.; Carbeck, J. D.; Rao, J.; Whitesides, G. M.
Electrophoresis 1998, 19, 367–382.
1
4
. Conclusions
It was shown that the enzymatic transformation of
carbohydrate-related native enzymes using the plug–
plug format under electrophoretically mediated micro-
analysis conditions was possible. Furthermore, the
kinetic analyses were achieved while the scale of the
enzymatic reactions was radically reduced (1000-fold
improvement when compared with the off-line analysis
of the reaction carried out in a microcentrifuge tube or
by a method using radiolabeled materials). Thus, we
were able to perform enzymatic reactions in a reaction
volume of a few nL, whereas a limit in reaction volume
for the traditional methods was ꢀ20 mL due to surface
tension. This is a dramatic improvement in methodolo-
gies, because a traditional assay, a photometric assay
for example, takes as much as 400 mL (4×20 mL/reac-
4. Yoshimoto, Y.; Shibukawa, A.; Sasagawa, H.; Nitta, S.;
Nakagawa, T. J. Pharm. Biomed. Anal. 1995, 13, 483–
488.
5
6
. Freitag, R. J. Chromatogr. B 1999, 722, 279–301.
. Bao, J. J.; Fujima, J. M.; Danielson, N. D. J. Chro-
matogr. B 1997, 699, 481–497.
7. Taga, A.; Nishino, A.; Honda, S. Chromatography 1997,
8, 320–321.
. Jin, Z.; Chen, R.; Colon, L. A. Anal. Chem. 1997, 69,
326–1331.
1
8
9
1
. Regnier, F. E.; Patterson, D. H.; Harmon, B. J. Trends
Anal. Chem. 1995, 14, 177–181.
10. Wu, D.; Regnier, F. E. Anal. Chem. 1993, 65, 2029–2035.
1
1
1
1. Chang, H.-T.; Yeung, E. S. Anal. Chem. 1993, 65, 2947–
951.
2. Avila, L. Z.; Whitesides, G. M. J. Org. Chem. 1993, 58,
508–5512.
2
tion×5 points) to obtain the K value, while EMMA
only requires a total of 20 mL. As a representative of
the EMMA, enzyme reactions of a-glucosidase, b-
m
5
3. Bao, J.; Regnier, F. E. J. Chromatogr. 1992, 608, 217–
224.

1
762
Y. Kanie, O. Kanie / Carbohydrate Research 337 (2002) 1757–1762
1
1
1
1
4. Zhao, D. S.; Gomez, F. A. Electrophoresis 1998, 19,
20–426.
5. Van Dyck, S.; Van Schepdael, A.; Hoogmartens, J. Elec-
trophoresis 2001, 22, 1436–1442.
6. Xu, Y.; Liu, X.; Ip, M. P. C. J. Liq. Chromatogr. Rel.
Technol. 1998, 21, 2781–2797.
7. Saevels, J.; Van Schepdael, A.; Hoogmartens, J. Elec-
trophoresis 1996, 17, 1222–1227.
21. Sun, X.-L.; Kanie, Y.; Guo, C.-T.; Kanie, O.; Suzuki, Y.;
4
Wong, C.-H. Eur. J. Org. Chem. 2000, 2643–2653.
22. Takebayashi, M.; Hiranuma, S.; Kanie, Y.; Kajimoto, T.;
Kanie, O.; Wong, C.-H. J. Org. Chem. 1999, 64, 5280–
5291.
23. Kanie, Y.; Kirsch, A.; Kanie, O.; Wong, C.-H. Anal.
Biochem. 1998, 263, 240–245.
24. Rassi, Z. E. Electrophoresis 1999, 20, 3134–3144.
25. Kanie, Y.; Kanie, O. Unpublished data that will be
reported elsewhere.
26. Landers, J. P. Appendix 2, Calculations of Practical Use.
In Handbook of Capillary Electrophoresis; Landers, J. P.,
Ed.; CRC: Boca Raton, FL, 1994; pp 617–618.
1
1
8. Varki, A. Glycobiology 1993, 3, 97–130.
9. Saotome, C.; Wong, C.-H.; Kanie, O. Chem. Biol. 2001,
8
, 1061–1071.
20. Saotome, C.; Kanie, Y.; Kanie, O.; Wong, C.-H. Bioorg.
Med. Chem. 2000, 8, 2249–2261.