Regioselectivity in β-galactosidase-catalyzed transglycosylation for the enzymatic assembly of D-galactosyl-D-mannose

Article abstract of DOI:10.1271/bbb.68.2086

The regioselectivity of β-galactosidase derived from Bacillus circulans ATCC 31382 (β-1,3-galactosidase) in transgalactosylation reactions using D-mannose as an acceptor was investigated. This D-mannose associated regioselectivity was found to be different from reactions using either GlcNAc or GalNAc as acceptors, not only for β-1,3-galactosidase but also for β-galactosidases of different origins. The relative hydrolysis rate of Galβ-pNP and D-galactosyl-D-mannoses, of various linkages, was also measured in the presence of β-1,3-galactosidase and was found to correlate well with the ratio of disaccharides formed by transglycosylation. The unexpected regioselectivity using D-mannose can therefore be explained by an anomalous specificity in the hydrolysis reaction. By utilizing the identified characteristics of both regioselectivity and hydrolysis specificity using D-mannose, an efficient method for enzymatic synthesis of β-1,3-, β-1,4- and β-1,6-linked D-galactosyl-D-mannose was subsequently established.

Full text of DOI:10.1271/bbb.68.2086

This article was downloaded by: [University of North Texas]
On: 29 November 2014, At: 20:29
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer
House, 37-41 Mortimer Street, London W1T 3JH, UK
Bioscience, Biotechnology, and Biochemistry
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/tbbb20
Regioselectivity in β-Galactosidase-catalyzed
Transglycosylation for the Enzymatic Assembly of D-
Galactosyl-D-mannose
Mariko MIYASATOa & Katsumi AJISAKAb
a Food Science Institute, Meiji Dairies Co.
b Niigata University of Pharmacy and Applied Life Sciences
Published online: 22 May 2014.
To cite this article: Mariko MIYASATO & Katsumi AJISAKA (2004) Regioselectivity in β-Galactosidase-catalyzed
Transglycosylation for the Enzymatic Assembly of D-Galactosyl-D-mannose, Bioscience, Biotechnology, and Biochemistry,
68:10, 2086-2090, DOI: 10.1271/bbb.68.2086
To link to this article: http://dx.doi.org/10.1271/bbb.68.2086
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of
the Content. Any opinions and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied
upon and should be independently verified with primary sources of information. Taylor and Francis shall
not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other
liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or
arising out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions

Biosci. Biotechnol. Biochem., 68 (10), 2086–2090, 2004
Regioselectivity in ꢀ-Galactosidase-catalyzed Transglycosylation
for the Enzymatic Assembly of D-Galactosyl-D-mannose
y
1
2;
Mariko MIYASATO and Katsumi AJISAKA
1Food Science Institute, Meiji Dairies Co., 540 Naruda, Odawara, Kanagawa 250-0862, Japan
2Niigata University of Pharmacy and Applied Life Sciences, 265-1 Higashijima, Niitsu, Niigata 956-8603, Japan
Received May 13, 2004; Accepted July 21, 2004
The regioselectivity of ꢀ-galactosidase derived from
Bacillus circulans ATCC 31382 (ꢀ-1,3-galactosidase) in
transgalactosylation reactions using D-mannose as an
acceptor was investigated. This D-mannose associated
regioselectivity was found to be different from reactions
using either GlcNAc or GalNAc as acceptors, not only
for ꢀ-1,3-galactosidase but also for ꢀ-galactosidases of
different origins. The relative hydrolysis rate of Galꢀ-
pNP and D-galactosyl-D-mannoses, of various linkages,
was also measured in the presence of ꢀ-1,3-galactosidase
and was found to correlate well with the ratio of
disaccharides formed by transglycosylation. The unex-
pected regioselectivity using D-mannose can therefore be
explained by an anomalous specificity in the hydrolysis
reaction. By utilizing the identified characteristics of
both regioselectivity and hydrolysis specificity using D-
mannose, an efficient method for enzymatic synthesis of
ꢀ-1,3-, ꢀ-1,4- and ꢀ-1,6-linked D-galactosyl-D-mannose
was subsequently established.
In the present study, we investigated the transglyco-
sylation regioselectivity of ꢀ-1,3-galactosidase from
B. circulans when the acceptor was changed from D-
galactose to D-mannose. Moreover, we examined similar
transglycosylation reactions using ꢀ-galactosidases from
various origins and compared the regioselectively of
these different reactions, using either GalNAc or D-
mannose acceptors. We discuss the reasons for the
altered regioselectivites due to the change of acceptor
from either GlcNAc or GalNAc to D-mannose. To our
knowledge, the synthesis of D-galactosyl-D-mannose
using glycosidases has not been reported previously.
We also explain an efficient method for the enzymatic
synthesis of ꢀ-1,3-, ꢀ-1,4- and ꢀ-1,6-linked D-galacto-
syl-D-mannoses.
Materials and Methods
Materials. ꢀ-Galactosidases from E. coli and from
bovine testes were purchased from Sigma. ꢀ-1,4-
Galactosidase from B. circulans was purchased from
Wako Pure Chemicals. (Osaka, Japan). ꢀ-Galactosidases
from Aspergillus oryzae and from Penicillium multi-
color were purchased from Toyobo (Osaka, Japan) and
K.I. Chemicals (Shizuoka, Japan) respectively. ꢀ-Ga-
lactosidase from Bifidobacterium bifidum and ꢀ-1,3-
galactosidase from B. circulans (recombinant) were
used without further purification of the culture broth.
Key words: ꢀ-galactosidase; Bacillus circulans; regio-
selectivity; transglycosylation; D-galactosyl-
D-mannose
Oligosaccharides are known to be synthesized regio-
selectively by transglycosylation using glycosidases, and
this holds true also for galactosyl oligosaccharides.
Many biologically important galactosyl oligosaccharides
have been synthesized enzymatically.1,2) A recombinant
ꢀ-galactosidase from Bacillus circulans ATCC 31382
(ꢀ-1,3-galactosidase)3) has been reported to have a
high specificity for both Galꢀ-(1!3)-GlcNAc and
Galꢀ(1!3)-GalNAc derivatives. When this enzyme
was used in transgalactosylation reactions using GlcNAc
and GalNAc as acceptors, it showed high regioselectiv-
ity in the formation of Galꢀ-(1!3)-GlcNAc and Galꢀ-
(1!3)-GalNAc respectively.4) In contrast, this trans-
glycosylation regioselectivity was found to change from
a ꢀ-(1!3)-linkage to a ꢀ-(1!6)-linkage by a change of
acceptor to D-galactose.5) Hence, in transglycosylation
reactions using D-galactose derivatives as acceptors,
Galꢀ-(1!6)-Gal derivatives were obtained exclusively.
Analytical methods. HPLC was carried out with an
HPAE-PAD Dionex PeakNet system using Carbopac
PA1 (4:0 ꢀ 250 mm) column eluted by either 40 or
1
64 mM sodium hydroxide solutions. H and 13C NMR
spectra were obtained using a Varian Inova 500
spectrometer.
Synthesis of Galꢀ-(1!3)-Man by ꢀ-1,3-galactosi-
dase. A reaction mixture consisting of Galꢀ-pNP
(150 mg), D-mannose (800 mg), and ꢀ-1,3-galactosidase
from B. circulans (2.0 U) in 2.1 ml of sodium phosphate
buffer (pH 6.0) containing 20% (v/v) DMF was incu-
bated at 37 ꢁC and the reaction was monitored by HPLC.
y
To whom correspondence should be addressed. Fax: +81-250-25-5021; E-mail: ajisaka@niigatayakudai.jp

ꢀ-Galactosidase-catalyzed Transgalactosylation to D-Mannose
2087
When the Galꢀ-pNP peak disappeared from the HPLC
spectrum, the reaction was stopped by heating the
solution for 5 min in boiling water. The reaction
produced a mixture of Galꢀ-(1!3)-Man and Galꢀ-
(1!6)-Man in an approximate ratio of 2:1. In our
typical experiment, the reactions were stopped at 2 h.
The reaction mixture was then diluted into 1 liter with
0.1 M sodium phosphate buffer (pH 6.0), and incubated
at 37 ꢁC in the presence of ꢀ-galactosidase from E. coli
(10 U). After overnight incubation, the solution was
applied onto an activated carbon column (2:7 ꢀ 50 cm).
The column was washed with 1 liter of water and
subsequently eluted with a gradient of water (1 liter)-
30% (v/v)-ethanol (1 liter). The sugar was detected with
the PAD-HPLC system using a sodium hydroxide
solution, and fractions containing disaccharides were
collected and concentrated in vacuo and freeze-dried to
give 15.4 mg of highly purified and 9.4 mg of moder-
ately pure Galꢀ-(1!3)-Man. NMR (D2O): ꢁ 3.61 (t, 1H,
J2;3 ¼ 7:62 Hz, H-2(Gal)), ꢁ 3.94 (bs, 1H, H-4(Gal)),
37 ꢁC. The reaction mixture contained Galꢀ-(1!3)-
Man, Galꢀ-(1!4)-Man, and Galꢀ-(1!6)-Man in an
approximately 10:4:86 ratio, detected by HPLC. The
unwanted Galꢀ-(1!3)-Man was hydrolyzed using ꢀ-
galactosidase from bovine testes (40 U). Galꢀ-(1!6)-
Man was purified using an activated carbon column by a
procedure similar to that described in the previous
sections. Highly purified (25.6 mg), moderately purified
(41.9 mg), and crude preparations (more than 30%
contaminants) (45.9 mg) of Galꢀ-(1!6)-Man were
obtained. NMR (D2O): ꢁ 3.44 (dd, 1H, J2;3 ¼ 9:91 Hz,
H-2(Galꢀ)), ꢁ 4.08 (dd, 1H, J5;6 ¼ 1:97 Hz, J6;6
¼
11:29 Hz, H-6(Gal)), ꢁ 4.32 (d, 1H, J1;2 ¼ 7:86 Hz, H-
1(Galꢀ)), ꢁ 4.33 (d, 1H, J1;2 ¼ 7:85 Hz, H-1(Galꢂ)), ꢁ
4.79 (d, 1H, J1;2 ¼ 0:74 Hz, H-1(Manꢀ)), ꢁ 5.05 (d, 1H,
J1;2 ¼ 1:97 Hz, H-1(Manꢂ)).
Synthesis of D-galactosyl-D-mannose using ꢀ-galac-
tosidase from various origins. A reaction mixture
consisting of Galꢀ-pNP (30 mg), D-mannose (160 mg),
and ꢀ-galactosidase (0.8 U) in 400 ꢃl of sodium phos-
phate buffer (pH 6.0) containing 20% (v/v) DMF was
incubated at 37 ꢁC, and the reaction was monitored by
HPLC. When the peak of Galꢀ-pNP was no longer
detectable by HPLC, the reaction was stopped by
heating the solution for 5 min in boiling water. The
reaction yield of each disaccharide was calculated from
the peak areas in HPLC chromatograms, and is summa-
rized in Table 1.
ꢁ
3.98 (dd, 1H, J3;4 ¼ 5:57 Hz, J3;4 ¼ 3:23 Hz,
H-3(Manꢀ)), ꢁ 4.06 (dd, 1H, J2;3 ¼ 2:93 Hz, J3;4
¼
6:46 Hz, H-3(Manꢂ)), ꢁ 4.13 (bs, 1H, H-2(Manꢂ)), ꢁ
4,15 (d, J2;3 ¼ 3:23 Hz, 1H, H-2(Manꢀ)), ꢁ 4.54 (d, 1H,
J1;2 ¼ 7:62 Hz, 1H, H-1(Galꢂ)), ꢁ 4.55 (d, 1H, J1;2
¼
7:92 Hz, H-1(Galꢀ)), ꢁ 4.91 (s, 1H, H-1(Manꢀ)), ꢁ 5.23
(d, 1H, J1;2 ¼ 0:88 Hz, H-1(Manꢂ)).
Synthesis of Galꢀ-(1!4)-Man by ꢀ-1,4-galactosidase
from B. circulans. A reaction mixture consisting of
Galꢀ-pNP (150 mg), D-mannose (800 mg), and ꢀ-1,4-
galactosidase from B. circulans (2.0 U) in sodium
phosphate buffer (2.1 ml, pH 6.0) containing 20% (v/
v) DMF was incubated at 37 ꢁC. The reaction produced a
mixture of Galꢀ-(1!3)-Man, Galꢀ-(1!4)-Man, and
Galꢀ-(1!6)-Man in an approximately 1:3:1 ratio. In
order to remove unwanted disaccharides, the hydrolysis
reaction was performed using ꢀ-1,3-galactosidase from
B. circulans (10 U), which was predicted to hydrolyze
Galꢀ-(1!3)-Man and Galꢀ-(1!6)-Man. The solution
containing Galꢀ-(1!4)-Man as a unique disaccharide
was applied onto an activated carbon column, and the
elution procedure was the same as that for the
purification of Galꢀ-(1!3)-Man. A small amount
(1.5 mg) of highly purified Galꢀ-(1!4)-Man was
obtained together with 28.7 mg of Galꢀ-(1!4)-Man
with lower purity. NMR (D2O): ꢁ 3.45 (dd, 1H,
Table 1. Synthesis of Transfer Products Mediated by Various ꢀ-
Galactosidases
Acceptor
Origin of enzyme
E. coli
Linkage
GalNAc
Man
ꢀ-(1!3)
ꢀ-(1!4)
ꢀ-(1!6)
trace
trace
95
10
4
86
A. Oryzae
ꢀ-(1!3)
ꢀ-(1!4)
ꢀ-(1!6)
5
3
17
12
71
92
P. multicolor
ꢀ-(1!3)
ꢀ-(1!4)
ꢀ-(1!6)
trace
6
25
30
45
94
ꢀ-1,4-galactosidase
from B. circulans
ꢀ-(1!3)
ꢀ-(1!4)
ꢀ-(1!6)
trace
63
20
62
18
37
J2;3 ¼ 8:06 Hz, H-2(Gal)), ꢁ 3.79 (dd, 1H, J2;3
¼
3:43 Hz, J3;4 ¼ 9:81 Hz, H-3(Manꢂ)), ꢁ 4.00 (bs, 1H,
H-2(Manꢂ)), ꢁ 4.34 (d, 1H, J1;2 ¼ 7:71 Hz, H-1(Gal)), ꢁ
4.47 (d, 1H, J2;3 ¼ 2:94 Hz, H-2(Manꢀ)), ꢁ 4.87 (s, 1H,
H-1(Manꢀ)), ꢁ 5.18 (s, 1H, H-1(Manꢂ)).
B. bifidum
ꢀ-(1!3)
ꢀ-(1!4)
ꢀ-(1!6)
trace
64
23
39
38
36
ꢀ-1,3-galactosidase
from B. circulans
ꢀ-(1!3)
ꢀ-(1!4)
ꢀ-(1!6)
95
trace
trace
67
trace
33
Synthesis of Galꢀ-(1!6)-Man by ꢀ-galactosidase
from E. coli. A reaction mixture consisting of Galꢀ-pNP
(600 mg), D-mannose (3.2 g), and ꢀ-galactosidase from
E. coli (8.0 U) in 0.1 M sodium phosphate buffer (8.0 ml,
pH 6.0) containing 20% (v/v) DMF was incubated at
Streptococcus 6646 K
ꢀ-(1!3)
ꢀ-(1!4)
ꢀ-(1!6)
trace
100
trace
100
trace
trace

2088
M. MIYASATO and K. AJISAKA
Measurement of relative hydrolysis rate of Galꢀ-pNP
and D-galactosyl-D-mannose disaccharides. 1.5 mM of
each of the disaccharides or of a Galꢀ-pNP solution in
240 ꢃl of 0.1 M sodium phosphate buffer (pH 6.0) and
20 ꢃl of 24 mM L-fucose solution were mixed with 10 ꢃl
of enzyme solution (5.0 units/ml) and incubated at
37 ꢁC. The ratio of either the remaining disaccharides or
Galꢀ-pNP was calculated from the HPLC peak areas. L-
Fucose was used as an internal standard.
Results and Discussion
Transgalactosylation to D-mannose using ꢀ-1,3-ga-
lactosidase
Zeng et al. reported that transgalactosylation with a D-
galactose acceptor yielded Galꢀ-(1!6)-Gal as the main
product.5) In the present study, we used D-mannose as an
acceptor to examine how ꢀ-galactosidase would dis-
criminate among the acceptor substrates in the trans-
glycosylation reactions. D-mannose has a stereochemis-
try largely different from those of D-galactose or D-
GalNAc, which are characterized by axial and equatorial
OH groups at the positions of C-2 and C-4 respectively.
Moreover, the anomeric OH group is fixed predomi-
nantly at the ꢂ-configuration, which simplifies the
identification of ꢀ-galactosylation products. The HPLC
chromatogram of the reaction mixture of Galꢀ-pNP and
D-mannose in the presence of ꢀ-1,3-galactosidase from
B. circulans is shown in Fig. 1. The chart shows two
disaccharide peaks representing Galꢀ-(1!3)-Man and
Galꢀ-(1!6)-Man with a ratio of approximately 2:1, but
peaks indicating the presence of Galꢀ-(1!2)-Man or
Galꢀ-(1!4)-Man were not detected.
Fig. 2. Time Courses of the Peak Areas of Galꢀ-pNP and D-
Galactosyl-D-mannose of Various Linkages in the Presence of ꢀ-1,3-
Galactosidase from B. circulans.
Galꢀ-pNP ( ), Galꢀ-(1!3)-Man ( ), Galꢀ-(1!6)-Man ( ),
Galꢀ-(1!4)-Man ( ).
Hydrolysis of D-galactosyl-D-mannoses containing
various different linkages by ꢀ-1,3-galactosidase
Generally, regioselectivity strongly correlates with
hydrolysis specificity,6,7) and hence we measured the
relative hydrolysis rates of D-galactosyl-D-mannose and
Galꢀ-pNP. The results, shown in Fig. 2, indicate that the
relative hydrolysis rate is in the following order: Galꢀ-
pNP > Galꢀ-(1!3)-Man > Galꢀ-(1!6)-Man > Galꢀ-
(1!4)-Man. This finding is consistent with the results
for the relative yield of the respective disaccharides by
transglycosylation. Galꢀ-(1!3)-Man was hydrolyzed
faster than Galꢀ-(1!6)-Man and was produced more
readily by transgalactosylation. In addition, Galꢀ-
(1!4)-Man was hydrolyzed slowly and was not
obtained by transglycosylation, this result is also con-
sistent with the hydrolysis specificity. Hence it may be
concluded that the regioselectivity in these transglyco-
sylation reactions is not anomalous, but due to the
hydrolysis specificity of D-galactosyl-D-mannose, which
is less restrictive than that of either Gal-GlcNAc or Gal-
GalNAc.
Regioselectivity in transgalactosylation by ꢀ-galacto-
sidases from various origins
We next examined reactions using other ꢀ-galactosi-
dases of various origins.8,9) The results are summarized
in Table 1. ꢀ-1,4-Galactosidase from B. circulans is
known to yield ꢀ-(1!4)-linked disaccharide predom-
inantly by transgalactosylation when either GlcNAc or
GalNAc is used as the acceptor.10–12) In our experiments,
however, we obtained yields of Galꢀ-(1!3)-Man,
Galꢀ-(1!4)-Man, and Galꢀ-(1!6)-Man in a ratio of
approximately 1:3:1 in reactions with this enzyme when
D-mannose was used as an acceptor. The regioselectivity
Fig. 1. HPAE-PAD Chromatography of the Reaction Mixture (3:
1.5 h, 2: 0.5 h, 1: 0 h) Using Galꢀ-pNP as a Donor and Man as an
Acceptor in the Presence of ꢀ-1,3-Galactosidase from B. circulans.
Column: Carbopac PA1, Elution: 40 mM sodium hydroxide
solution. Flow rate: 1 ml/min, peak A: D-galactose and D-mannose,
peak B: Galꢀ-pNP, peak C: Galꢀ-(1!6)-Man, peak D: Galꢀ-
(1!3)-Man.

ꢀ-Galactosidase-catalyzed Transgalactosylation to D-Mannose
2089
changed due to the structural change in the acceptor
molecule. In reactions using ꢀ-galactosidase from
E. coli, the disaccharides that we obtained were Galꢀ-
(1!3)-Man, Galꢀ-(1!4)-Man, and Galꢀ-(1!6)-Man,
in a ratio of 10:4:86. This enzyme is known to yield
Galꢀ-(1!6)-GalNAc exclusively in reactions using
Galꢀ-pNP and GalNAc.13) In this case also the regio-
selectivity of the reactions weakened slightly. It is
noteworthy that ꢀ-galactosidase from Streptococcus
6646 K showed high regioselectivity for ꢀ-1,4-linkages,
similarly to reactions using GlcNAc or GalNAc as an
acceptor.14) As can be seen in Table 1, the additional
enzyme types showed different regioselectivities to the
reactions using GlcNAc or GalNAc as an acceptor.
Hence, the so-called ‘‘regioselectivity’’ is a regioselec-
tivity in transglycosylation to either GlcNAc or GalNAc.
Consequently, in transgalactosylation to D-galactose or
D-mannose, one should bear in mind that there are
alterations to or even a complete loss of established
regioselectivities.
B. circulans, Galꢀ-(1!3)-Man was mainly produced
with Galꢀ-(1!6)-Man as a minor component. In this
reaction, Galꢀ-(1!4)-Man levels were negligibly small.
Galꢀ-(1!6)-Man was then hydrolyzed using ꢀ-galac-
tosidase from E. coli to generate a mixture of mono-
saccharides and Galꢀ-(1!3)-Man, which subsequently
could easily be isolated using activated carbon column
chromatography. Similarly, Galꢀ-(1!4)-Man was ob-
tained as follows: A mixture of Galꢀ-(1!3)-Man, Galꢀ-
(1!4)-Man, and Galꢀ-(1!6)-Man generated by ꢀ-1,4-
galactosidase from B. circulans was treated by ꢀ-1,3-
galactosidase from B. circulans, which has hydrolysis
specificity for ꢀ-(1!3)- and ꢀ-(1!6)-linkages. Galꢀ-
(1!3)-Man and Galꢀ-(1!6)-Man were therefore de-
pleted by hydrolysis, leaving Galꢀ-(1!4)-Man as a
remainder. The procedure for obtaining Galꢀ-(1!6)-
Man is very simple, since ꢀ-galactosidase from E. coli
generates a mixture of Galꢀ-(1!3)-Man and Galꢀ-
(1!6)-Man in a ratio of approximately 1:9, together
with small amounts of Galꢀ-(1!4)-Man. Galꢀ-(1!3)-
Man is then easily removed by hydrolysis using ꢀ-
galactosidase from bovine testes.15,16) The remaining
small amount of Galꢀ-(1!4)-Man can then be removed
during the activated carbon column chromatography
step. The results are summarized in Fig. 3. The
13C NMR data of the purified D-galactosyl-D-mannoses
are summarized in Table 2.
Preparative study of D-galactosyl-D-mannoses of
differing linkages
Since regioselectivity in transgalactosylation to D-
mannose is not as rigid as in the case of transgalacto-
sylation to GlcNAc or GalNAc, the enzymatic prepara-
tion of D-galactosyl-D-mannose is somewhat more
cumbersome. Galꢀ-(1!3)-Man was obtained as fol-
lows: First, by transglycosylation using Galꢀ-pNP and
D-mannose in the presence of ꢀ-1,3-galactosidase from
By combining transglycosylation with specific hy-
drolysis reactions, then, each disaccharide can be
isolated and purified. Thus synthesized D-galactosyl-D-
Fig. 3. Synthesis and Purification of D-Galactosyl-D-mannose of Various Linkages.

2090
M. MIYASATO and K. AJISAKA
Table 2. 13C NMR Data for D-Galactosyl-D-mannose
7) Ajisaka, K., Fujimoto, H., and Miyasato, M., An ꢂ-L-
fucosidase from Penicillium multicolor as a candidate
enzyme for the synthesis of ꢂ(1!3)-linked fucosyl
oligosaccharides by transglycosylation. Carbohydr. Res.,
309, 125–129 (1998).
Compound
Gal-ꢀ-(1!3)Man Gal-ꢀ-(1!4)Man Gal-ꢀ-(1!6)Man
Gal C1
C2
C3
101.826
71.672
73.457
69.632
76.147
61.946
94.581
69.214
78.915
66.103
73.457
61.779
102.432
69.703
70.749
67.223
75.407
61.384
92.410
68.830
72.538
78.238
72.640
60.678
103.531
70.808
70.970
68.830
75.348
61.204
94.311
70.324
71.545
66.788
72.797
69.016
8) Yoon, J. H., and Ajisaka, K., The synthesis of galacto-
pyranosyl derivatives with ꢀ-galactosidases of different
origins. Carbohydr. Res., 292, 153–163 (1996).
9) Dumorter, V., Montreuil, J., and Bouquelet, S., Primary
structure of ten galactosides formed by transglycosyla-
tion during lactose hydrolysis by Bifidobacterium bifi-
dum. Carbohydr. Res., 201, 115–123 (1990).
10) Sakai, K., Katsumi, R., Ohi, H., Usui, T., and Ishido, Y.,
Enzymatic syntheses of N-acetyllactosamine and N-
acetylallolactosamine by the use of ꢀ-D-galactosidases.
J. Carbohydr. Chem., 11, 553–565 (1992).
C4
C5
C6
Man C1
C2
C3
C4
C5
C6
11) Usui, T., Kubota, S., and Ohi, H., A convenient synthesis
of ꢀ-galactosyl disaccharide derivatives using the ꢀ-D-
galactosidase from Bacillus circulans. Carbohydr. Res.,
244, 315–323 (1993).
12) Usui, T., Morimoto, S., Hayakawa, Y., Kawaguchi, M.,
Murata, T., Matahira, Y., and Nishida, Y., Regioselec-
tivity of ꢀ-galactosyl-disaccharide formation using the
ꢀ-D-galactosidase from Bacillus circulans. Carbohydr.
Res., 285, 29–39 (1996).
13) Hedbys, L., Larsson, P. O., Mosbach, K., and Svensson,
S., Synthesis of the disaccharide 6-O-ꢀ-D-galactopyra-
nosyl-2-acetamideo-2-deoxy-D-galactose using immobi-
lized ꢀ-galactosidase. Biochem. Biophys. Res. Commun.,
123, 8–15 (1984).
mannoses of different linkages may be utilized for
various biological studies. For example, a repeating unit
of D-galactosyl-D-mannose has been found to be a
component unit of Leishmania lipophosphoglycan, and
the disaccharide unit is supposed to be related to
endothelial adhesion to monocyte.17,18) Therefore these
disaccharides are expected to be utilized to clarify the
biological roles and functions of D-galactosyl-D-man-
nose sequence in Leishmania.
References
14) Kiyohara, T., Terao, T., Nakana, K. S., and Osawa,
T., Purification and characterization of ꢀ-N-acetyl-
hexosaminidases and ꢀ-galactosidase from Streptococ-
cus 6646 K. J. Biochem., 80, 9–17 (1976).
15) Hedbys, L., Johansson, E., Mosbach, K., Larsson, P. O.,
Gunnarsson, A., Svensson, S., and Lonn, H., Synthesis of
Galꢀ1-3GlcNAc and Galꢀ1-3GlcNAcꢀ-SEt by an enzy-
matic method comprising the sequential use of ꢀ-
galactosidases from bovine testes and Escherichia coli.
Glycoconjugate J., 6, 161–168 (1989).
1) Matsuo, I., Isomura, M., and Ajisaka, K., The first
synthesis of Neu5Acꢂ2-3Galꢀ1-4GlcNAcꢀ1-2Manꢂ1-
Ser — a new discovered component of ꢂ-dystroglycan.
Tetrahedron Lett., 40, 5047–5050 (1999).
2) Ajisaka, K., Miyasato, M., Ito, C., Fujita, F., Yamazaki,
Y., and Oka, S., Linkage of sugar chains to a fragment
peptide of FGF-5S by a chemoenzymatic strategy and
changes in the rate of proteolytic hydrolysis. Glycocon-
jugate J., 18, 301–308 (2001).
3) Ito, Y., and Sasaki, T., Cloning and characterization of
the gene encoding a novel ꢀ-galactosidase from Bacillus
circulans. Biosci. Biotechnol. Biochem., 61, 1270–1276
(1997).
4) Fujimoto, H., Miyasato, M., Ito, Y., Sasaki, T., and
Ajisaka, K., Purification and properties of recombinant
ꢀ-galactosidase from Bacillus circulans. Glycoconjugate
J., 15, 155–160 (1998).
5) Zeng, X., Yoshino, R., Murata, T., Ajisaka, K., and Usui,
T., Regioselective synthesis of p-nitrophenyl glycosides
of ꢀ-D-galactopyranosyl disaccharides by transglycosy-
lation with ꢀ-galactosidases. Carbohydr. Res., 325, 120–
131 (2000).
6) Ajisaka, K., and Yamamoto, Y., Control of the regio-
selectivity in the enzymatic synthesis of oligosaccharides
using glycosidases. Trends in Glycosci. Glycotech., 14,
1–11 (2002).
16) Gambert, U., Gonzalez, R., Farkas, E., Thiem, J.,
Bencomo, V. V., and Liptak, A., Galactosylation with
ꢀ-galactosidase from bovine testes employing modified
acceptor substrates. Bioorg. Med. Chem., 5, 1285–1291
(1997).
17) Hatzigeorgiou, D. E., Geng, J., Zhu, B., Zhang, Y., Liu,
K., Rom, W. N., Fenton, M. J., Turco, S. J., and Ho, L.,
Lipophosphoglycan from Leishmania suppresses ago-
nist-induced interleukin 1ꢀ gene expression in human
monocytes via a unique promoter sequence. Proc. Natl.
Acad. Sci. U.S.A., 93, 14708–14713 (1996).
18) Lo, S. K., Bovis, L., Matura, R., Zhu, B., He, S., Lum,
H., Turco, S. J., and Ho, J. L., Leishmania Lipophos-
phoglycan reduces monocyte transendotherial migration:
modulation of cell adhesion molecules, intercellular
junctional proteins, and chemoattractants. J. Immunol.,
160, 1857–1865 (1998).