Mode of action of a β-(1→6)-glucanase from Penicillium multicolor

Article abstract of DOI:10.1016/j.carres.2012.11.002

β-(1→6)-Glucanase from the culture filtrate of Penicillium multicolor LAM7153 was purified by ammonium sulfate precipitation, followed by cation-exchange and affinity chromatography using gentiotetraose (Gen ₄) as ligand. The hydrolytic mode of action of the purified protein on β-(1→6)-glucan (pustulan) was elucidated in real time during the reaction by HPAEC-PAD analysis. Gentiooligosaccharides (DP 2-9, Gen _2-9), methyl β-gentiooligosides (DP 2-6, Gen_2-6 β-OMe), and p-nitrophenyl β-gentiooligosides (DP 2-6, Gen _2-6 β-pNP) were used as substrates to provide analytical insight into how the cleavage of pustulan (DP? 320) is actually achieved by the enzyme. The enzyme was shown to completely hydrolyze pustulan in three steps as follows. In the initial stage, the enzyme quickly cleaved the glucan with a pattern resembling an endo-hydrolase to produce a short-chain glucan (DP? 45) as an intermediate. In the midterm stage, the resulting short-chain glucan was further cleaved into two fractions corresponding to DP 15-7 and DP 2-4 with great regularity. In the final stage, the lower oligomers corresponding to DP 3 and DP 4 were very slowly hydrolyzed into glucose and gentiobiose (Gen ₂). As a result, the hydrolytic cooperation of both an endo-type and saccharifying-type reaction by a single enzyme, which plays a bifunctional role, led to complete hydrolysis of the glucan. Thus, β-(1→6)-glucanase varies its mode of action depending on the chain length derived from the glucan.

Full text of DOI:10.1016/j.carres.2012.11.002

Carbohydrate Research 366 (2013) 6–16
Contents lists available at SciVerse ScienceDirect
Carbohydrate Research
journal homepage: www.elsevier.com/locate/carres
Mode of action of a b-(1?6)-glucanase from Penicillium multicolor
Takeshi Hattori a, Yasuna Kato b, Shuji Uno b, Taichi Usui a,b,
⇑
a Department of Bioscience, Graduate School of Science and Technology, Shizuoka University, Ohya 836, Suruga ward, Shizuoka 422-8529, Japan
b Department of Applied Biological Chemistry, Faculty of Agriculture, Shizuoka University, Ohya 836, Suruga ward, Shizuoka 422-8529, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
b-(1?6)-Glucanase from the culture filtrate of Penicillium multicolor LAM7153 was purified by ammo-
nium sulfate precipitation, followed by cation-exchange and affinity chromatography using gentiotetra-
ose (Gen4) as ligand. The hydrolytic mode of action of the purified protein on b-(1?6)-glucan (pustulan)
was elucidated in real time during the reaction by HPAEC-PAD analysis. Gentiooligosaccharides (DP 2-9,
Gen2-9), methyl b-gentiooligosides (DP 2-6, Gen2-6 b-OMe), and p-nitrophenyl b-gentiooligosides (DP 2-6,
Received 16 September 2012
Received in revised form 5 November 2012
Accepted 6 November 2012
Available online 15 November 2012
ꢀ
Gen2-6 b-pNP) were used as substrates to provide analytical insight into how the cleavage of pustulan (DP
Keywords:
320) is actually achieved by the enzyme. The enzyme was shown to completely hydrolyze pustulan in
three steps as follows. In the initial stage, the enzyme quickly cleaved the glucan with a pattern resem-
b-(1?6)-Glucanase
Hydrolytic manner
b-(1?6)-Glucan
Gentiooligosaccharide
Transglycosylation
ꢀ
bling an endo-hydrolase to produce a short-chain glucan (DP 45) as an intermediate. In the midterm
stage, the resulting short-chain glucan was further cleaved into two fractions corresponding to DP
15-7 and DP 2-4 with great regularity. In the final stage, the lower oligomers corresponding to DP 3
and DP 4 were very slowly hydrolyzed into glucose and gentiobiose (Gen2). As a result, the hydrolytic
cooperation of both an endo-type and saccharifying-type reaction by a single enzyme, which plays a
bifunctional role, led to complete hydrolysis of the glucan. Thus, b-(1?6)-glucanase varies its mode of
action depending on the chain length derived from the glucan.
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
(by an endo-type enzyme) and saccharification (by an exo-type
enzyme).23 During the first step, starch is solubilized in water
b-(1?6)-Glucan, which consists of glucose residues joined in a
b-(1?6)-linkage, is a structurally important component of the cell
wall of yeasts and fungi. It is produced in budding yeasts as a link
between cell wall proteins and the main b-(1?3)-glucan/chitin
polysaccharides.1 However, the polymer is much less abundant
in nature than other b-glucans, such as b-(1?4)- or b-(1?3)-glu-
can. Pustulan from Umbilicaria papullosa is well known as a typical
linear b-(1?6)-glucan.2 On the other hand, b-(1?6)-glucanase (EC
3.2.1.75) belongs to glycosyl hydrolase families 5 and 30 (http://
www.cazy.org/).3–6 Although b-(1?6)-glucanases are widely
distributed among fungi,7 few of them have been purified and
characterized.8–17 As a result, there are a few studies on the hydro-
lytic mode of action of b-(1?6)-glucanase on b-(1?6)-glucan.18–20
In general, enzymatic degradation of polysaccharides is considered
to consist of endo-acting enzymes, which randomly hydrolyze gly-
cosidic linkages in the polymers, and processive exo-acting en-
zymes, which degrade the polysaccharides from the polymer
ends. For example, cellulose21 and chitin22 are hydrolyzed by the
synergistic action of corresponding endo- and exo-acting enzymes.
In starch hydrolysis, the reaction is performed through liquefaction
and partially hydrolyzed with an a-amylase. In the second step,
saccharifying enzymes such as b-amylase or glucoamylase trans-
form the liquefied starch into final products that can be specific oli-
gosaccharides such as glucose or maltose.
The present paper describes the hydrolytic mode of action of
b-(1?6)-glucanase on pustulan based on kinetic and product anal-
ysis, together with specific purification of the enzyme and the
methods for synthesizing a series of gentiooligosides using its
transglycosylation activity.
2. Results
2.1. Purification of b-(1?6)-glucanase based on gentiotetraose
(Gen4)-immobilized affinity chromatography
To purify Penicillium multicolor b-(1?6)-glucanase, the crude
enzyme preparation was fractionated by ammonium sulfate pre-
cipitation. The desalted fraction was applied to cation-exchange
chromatography using CM-Sepharose Fast Flow, yielding a single
peak of b-(1?6)-glucanase activity that was completely devoid
of b-glucosidase activity. This fraction was then subjected to sub-
strate affinity chromatography on a Gen4-coupled TOYOPEARL
AF-Amino-650M column (Fig. 1A). b-(1?6)-Glucanase activity
⇑
Corresponding author. Tel.: +81 54 238 4305; fax: +81 54 238 8448.
E-mail address: usui@ipc.shizuoka.ac.jp (T. Usui).
0008-6215/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.carres.2012.11.002

T. Hattori et al. / Carbohydrate Research 366 (2013) 6–16
7
A
B
OH
O
HO
HO
O
OH
HO
O
HO
O
OH
O
HO
HO
O
OH
OH
H
N
OH
HO
O
HO
OH
Figure 1. (A) Structure of the Gen4-coupled affinity gel. (B) SDS–PAGE of purified P. multicolor b-(1?6)-glucanase using a 12.5% gel under reducing conditions. Proteins were
stained with CBB. Lane 1, crude enzyme; Lane 2, ammonium sulfate precipitation; Lane 3, cation-exchange elute; Lane 4, affinity elute. Numbers on the right indicate the
molecular mass of the protein standards. (Lane M).
Table 1
Summary of the purification and yield of b-(1?6)-glucanase from P. multicolor
Purification step
Total protein (mg)
Total activity (U)
Specific activity (U/mg)
Yield (%)
Purification (fold)
Crude enzyme
73.0
29.7
0.8
49.1
33.3
3.7
0.7
1.1
4.5
164.1
100.0
67.8
7.6
1.0
1.7
6.7
244.3
(NH4)2SO4 precipitation
CM Sepharose F.F.
Affinity
0.007
1.1
2.2
was strongly adsorbed on the column, and was subsequently easily
eluted as a single peak by a salt gradient. The fraction was highly
purified (244-fold relative to the crude enzyme preparation) with
a specific activity of 164.1 U/mg (Table 1). It gave a single protein
band by SDS–PAGE with a molecular mass of 51 kDa (Fig. 1B).
Furthermore, the protein was analyzed by MALDI-TOF mass spec-
trometry, and a peak with an m/z value of 49,909 was observed.
This suggests that the enzyme does take a monomer but not a unit
structure. The isoelectric point (pI) was not determined because it
was beyond the limit of detection (pI <3.5) (data not shown). The
optimum temperature for the purified enzyme was 50 °C, and heat
inactivation occurred above 60 °C. The optimum pH was pH 4.0,
and more than 90% of the maximum enzyme activity was main-
tained in the pH range 2.0–9.0 at 4 °C and 24 h of incubation.
The hydrolytic activity of the enzyme was prominent only for
pustulan (a linear b-(1?6)-glucan) (Table 2). Little activity was
detected toward soluble starch, carboxymethyl cellulose, or
curdlan, enabling us to conclude that the enzyme is a specific b-
(1?6)-glucanase.
Table 2
Substrate specificity of purified b-(1?6)-glucanase toward selected glucans and
oligosaccharides
Substrate
Linkage type
Activity (U/mg)
Pustulan
Yeast glucan
Soluble starch
Carboxymethyl cellulose
Curdlan (Alcaligenes faecalis)
Gen8
b-1,6
b-1,3:b-1,6
a-1,4:a-1,6
b-1,4
b-1,3
b-1,6
164.1
1.4
0
0
0
4.2
utilizing the above-purified b-(1?6)-glucanase. The transglycosy-
lation reaction was carried out at a high substrate concentration
of Gen2 b-OMe (4 M), which served as both the donor and acceptor.
After terminating the enzyme reaction, the reaction mixture was
subjected to pyridylamination with 2-amino-pyridine to give pyr-
idylaminated gentiooligosyl derivatives because it proved difficult
to separate the desired methyl b-gentiooligosides from gentiooli-
gosaccharide by-products by chromatography. The reaction solu-
tion was applied to TOYOPEARL HW-40S chromatography to
remove the excess of 2-aminopyridine. The fractions that showed
color development with phenol-sulfuric acid were further sub-
jected to charcoal-Celite chromatography with a linear gradient
of ethanol (Fig. 2A and Scheme 1). The chromatogram showed six
fractions (MG1–MG6), numbered according to their order of elu-
tion. Fractions MG1 and MG2 were recovered as Glc b-OMe and
Gen2 b-OMe, respectively. Fractions MG3–MG6 were obtained as
the desired products Gen3 b-OMe, Gen4 b-OMe, Gen5 b-OMe, and
Gen6 b-OMe in yields of 7.0%, 4.3%, 1.5%, and 0.8%, respectively.
The structures of the synthesized products were analyzed by 1H,
13C NMR (Table 3), and ESI-MS.
2.2. Synthesis of methyl b-gentiooligosides and p-nitrophenyl b-
gentiooligosides
A series of methyl b-gentiooligosides (DP 2-6, Gen2-6 b-OMe)
and p-nitrophenyl b-gentiooligosides (DP 2-6, Gen2-6 b-pNP) were
prepared as control samples for analyzing the hydrolytic mecha-
nism of the purified b-(1?6)-glucanase on the b-(1?6)-glucan
linkage. We have recently reported that gentiooligosaccharides
(DP 3-9) can be prepared from the starting substance Gen2 by
the sequential transglycosylation reactions of b-glucosidase and
b-(1?6)-glucanase.24 By following the reported methodology, a
series of methyl b-gentiooligosides (DP 2-6) were enzymatically
prepared in a similar way.
A series of p-nitrophenyl b-gentiooligosides (DP 2-6) were pre-
pared in a similar way to the methyl b-gentiooligosides. p-Nitro-
Gen2 b-OMe was first synthesized from methyl b-D-glucoside
(Glc b-OMe), which serves as both the donor and acceptor sub-
strate, through the transglycosylation reaction of Hypocrea jecorina
b-glucosidase. The resulting Gen2 b-OMe was then used as a start-
ing substance for obtaining methyl b-gentiooligosides (DP 3-6),
phenyl b-D-glucoside (Glc b-pNP) was used as a starting material
for the synthesis of Gen2 b-pNP. The reaction was performed by
using the transglycosylation activity of H. jecorina b-glucosidase
with a high monomer concentration (100 mM). The reaction

8
T. Hattori et al. / Carbohydrate Research 366 (2013) 6–16
B
A
PG4
MG3
MG4
1.0
40
30
20
10
0
MG1
MG2
PG2
3.0
2.0
1.0
0.0
PG5
PG3
0.5
0.0
MG5
MG6
PG6
0
30
60
90
120
150
180
0
50
100
150
Fraction number (50 ml each)
Fraction number (25 ml each)
Figure 2. (A) Charcoal-Celite chromatography of transglycosylation products formed by the action of b-(1?6)-glucanase on methyl b-gentiobioside. The elution positions of
MG1 and MG2–MG6 correspond to methyl b- -glucoside and methyl b-gentiooligosides (DP 2-6), respectively. The straight line indicates ethanol concentration (%). (B)
D
TOYOPEARL HW-40S chromatography of transglycosylation products formed by the action of b-(1?6)-glucanase on gentiotriose and p-nitrophenyl b-gentiobioside. The
elution positions of PG2–PG6 correspond to p-nitrophenyl b-gentiooligosides (DP 2-6), respectively.
mixture was applied to TOYOPEARL HW-40S chromatography to
obtain the target dimer in 15.2% yield. The resulting Gen2 b-pNP
was used for the synthesis of Gen3-6 b-pNP. The gentiooligosides
(DP 3-6) were synthesized by a b-(1?6)-glucanase-catalyzed
transglycosylation reaction from the Gen3 donor to a Gen2 b-pNP
acceptor. In this case, low temperature (4 °C) was effective for rais-
ing the transglycosylation rate (data not shown). The reaction mix-
ture was applied to TOYOPEARL HW-40S chromatography. The
chromatogram showed five peaks designated PG2–PG6 (Fig. 2B).
Fraction PG2 was recovered as Gen2 b-pNP (57.4% recovery). Fractions
PG3 and PG4 were obtained as the target products Gen3b-pNP and
Gen4 b-pNP in yields of 2.1% and 14.2%, respectively. Fractions PG5
and PG6 were each further purified by reverse-phase ODS chroma-
tography to remove the gentiooligosaccharide by-products. The
target products Gen5 b-pNP and Gen6 b-pNP were obtained in
yields of 2.6% and 1.7%, respectively. The structures of the synthe-
sized products were identified by 1H, 13C NMR (Table 4), and
ESI-MS analyses. The synthetic gentiooligosaccharides and genti-
ooligosides were used to analyze the mechanism of action of
b-(1?6)-glucanase on b-(1?6)-glucan as described below.
b-OMe: the major cleavages occurred at bonds 2, 3, and 4 in
approximately equal proportion. It should be noted that the non-
reducing end glucosyl residues of Gen4-6 b-OMe were either not
or scarcely cleaved by the enzyme. Gen3 b-OMe was hydrolyzed
slowly to form Glc b-OMe and Gen2 b-OMe in the ratio 91:9. The
presence of a methyl group at the reducing end did not seem to
prohibit the reaction. These results indicate that methyl b-genti-
ooligosides are cleaved through the terminal glycoside, and the
number of the susceptible bond varies with the chain-length of
substrate. Thus, the cleavage gradually shifts from bond 2 to bonds
3 and 4 with increasing DP. The cleavage frequencies of p-nitro-
phenyl b-gentiooligosides (DP 2-6) were similarly analyzed by
HPLC. For Gen4 b-pNP, the third bond (bond 3) from the p-nitro-
phenyl group was cut preferentially. The major cleavage of Gen5
b-pNP occurred at bonds 3 and 4, whereas bond 2 was not com-
pletely cleaved. The hydrolytic pattern of Gen6 b-pNP, which was
found to be the best substrate for the enzyme, was substantially
similar to that of Gen5 b-pNP. Notably, glucose was not detected
during any of the reactions with Gen4-6 b-pNP. This indicates that
the non-reducing-end glucosyl terminus of the derivatives with
DP 4 or higher is not subject to hydrolysis. The purified b-(1?6)-
glucanase acted slightly on Gen3 b-pNP (2% hydrolytic activity
relative to 100% for Gen6 b-pNP), which was a poor substrate. By
contrast, the cleavage point of derivatives with DP 4 or higher
shifted such that the enzyme kept away from the p-nitrophenyl
group at the terminus. In all the tested substrates, release of
p-nitrophenol was not observed by the HPLC-based assay and the
result also confirmed colorimetrically by a standard end-point as-
say (data not shown). As a result, the relative enzyme activity for
Gen4-6 b-pNP markedly increased in comparison to that for Gen3
b-pNP.
2.3. Hydrolysis of methyl b-gentiooligosides, p-nitrophenyl
b-gentiooligosides, and gentiooligosaccharides: bond cleavage
frequency
The mode of action of the purified b-(1?6)-glucanase on the
gentiooligosides and gentiooligosaccharides prepared above were
examined. The cleavage frequencies and relative hydrolytic rate
of the enzyme were measured on the basis of the initial velocity
of the hydrolytic reaction and are summarized in Tables 5 and 6.
2.3.1. Methyl b-gentiooligosides and p-nitrophenyl
b-gentiooligosides
2.3.2. Gentiooligosaccharides
The cleavage frequencies and relative hydrolytic rate of the
purified b-(1?6)-glucanase on methyl b-gentiooligosides (DP
2-6) were determined by high-pH anion-exchange chromatogra-
phy with pulsed amperometric detection (HPAEC-PAD) analysis
(Table 5). As compared with Gen3 b-OMe, Gen4-6 b-OMe acted as
fairly good substrates. For Gen3 b-OMe, the second glucosidic bond
(bond 2) from the methyl group was cut preferentially. The major
cleavage of Gen4 b-OMe occurred at the second (bond 2) and third
(bond 3) glucosidic bonds from the methyl group. The hydrolytic
pattern of Gen5 b-OMe was substantially similar to that of Gen6
The enzymatic cleavage of gentiooligosaccharides (DP 2-6) and
the relative rate of hydrolysis by the purified enzyme were ana-
lyzed by HPAEC-PAD. However, the position of cleavage in the
reducing oligosaccharides was not localized from the hydrolysate
of the oligomers, because of the lack of aglycon. Taking into
account the data for b-gentiooligosides, the frequencies of cleavage
of gentiooligosaccharides with DP 4 or higher were proposed on
the basis of the assumption that the glucosyl residue at the
non-reducing end was not hydrolyzed by the enzyme (Table 6).
The major cleavage of Gen4 occurred at the first (bond 1) and

T. Hattori et al. / Carbohydrate Research 366 (2013) 6–16
9
OH
O
HO
HO
O
OH
HO
O
OCH3
HO
OH
β-(1 6)-glucanase
OH
O
HO
HO
O
OH
O
HO
HO
O
OH
n
O
HO
OCH3
HO
n=1- 4
OH
OH
O
HO
HO
O
OH
O
HO
HO
O
m
HO
OH
O
HO
OH
OH
Pyridylamination
OH
O
HO
HO
O
OH
O
HO
HO
O
HO
OH
n
O
HO
HO
OCH3
n=1- 4
OH
OH
O
HO
HO
O
OH
O
O
m
HO
HO
OH
OH
N
OH
OH
H
N
O
HO
HO
O
HO
OH
O
HO
HO
O
TOYOPEARL HW-40S
Charcoal-Celite
OH
m
N
OH
OH
H
N
HO
HO
OH
O
HO
O
HO
OH
O
HO
HO
O
OH
n
O
HO
OCH3
HO
n=1- 4
OH
Scheme 1. Synthesis and purification of methyl b-gentiooligosides.
second (bond 2) glucosidic bonds from the reducing terminus, and
the activity was much lower relative to that of Gen5, suggesting
that gentiooligomers of DP 5 and higher are good substrates.
Regarding Gen5 and Gen6, there was less cleavage at bond 1 as
the DP increased, with cleavage shifting to bonds 2 and 3/4, respec-
tively. Gen2 was found to act only slightly as a substrate.
sis, no signal was observed (Fig. 3A). During the early stage of
hydrolysis, doublets at 4.66–4.68 ppm (J = 8 Hz), characteristic of
H-1b from gentiooligosaccharides with different chain lengths, ap-
peared and rapidly increased in intensity over time (Fig. 3B and C).
Instead of these signals, a small doublet at 5.25 ppm (J = 4 Hz),
characteristic of H-1a from the corresponding gentiooligosaccha-
rides, appeared at 1 h after addition of the enzyme and gradually
increased over time (Fig. 3C–E). These observations suggest that
the b-anomer is initially formed and then easily converted into
2.4. Hydrolytic mechanism of b-(1?6)-glucanase on pustulan
2.4.1. NMR analysis
the a-anomer by mutarotation.
To elucidate the mechanism of pustulan hydrolysis by the en-
zyme, the stereochemistry of the hydrolysis was investigated by
1H NMR analysis. Figure 3 shows spectra of the time course of
the hydrolysis reaction in the 4.6–5.3 ppm region. Before hydroly-
2.4.2. HPAEC-PAD analysis
The manner of the hydrolysis of b-(1?6)-glucan (pustulan) by
the purified b-(1?6)-glucanase was analyzed in real time by

10
T. Hattori et al. / Carbohydrate Research 366 (2013) 6–16
Table 3
glucan (Fig 4B). After 10 min of incubation, the peak corresponding
13C-chemical shifts of methyl b-gentiooligosides (DP 2-6) in D2O solution (30 °C)
ꢀ
ꢀ
to DP 45 had decreased to afford a main broad peak of DP 15 with
minor peaks of DP 2-4 (Fig. 4C). After 30 min of incubation, the
Compound
Residuea
Chemical shift (d)
C-3 C-4 C-5
ꢀ
resulting oligomer (DP 15) was further distributed into two peak
C-1
C-2
C-6
CH3
ꢀ
groups corresponding to DP 10 and DP 2-4 (Fig. 4D). One should
Gen2 b-OMe CH3
60.2
remark that the peak of DP 2-4 increased in proportion with great
regularity over time, while a low level of glucose was detected dur-
1
2
106.2
105.7
75.87 78.5
75.91 78.5
72.3 77.7
72.5 78.7
71.4
63.6
ꢀ
ing the reaction. The resulting oligomers (DP 10) were slowly de-
Gen3 b-OMe CH3
60.2
60.2
ꢀ
graded over time to two peak groups corresponding to DP 7 and
1
2
3
106.2
105.74 75.87 78.4
105.69 75.93 78.5
75.87 78.5
72.3 77.7
72.2 77.8
72.5 78.7
71.6
71.4
63.6
DP 2-4 after 2 h of incubation (Fig. 4E). In the final stage of the
reaction, the cleavage of oligomers (DP 3-4) into glucose and
Gen2 proceeded extremely slowly, and it took prolonged incuba-
tion (96 h) to reach this stage (Fig. 4F).
Gen4 b-OMe CH3
1
2
3
4
106.2
105.7
105.7
105.8
75.88 78.5
75.88 78.4
75.88 78.4
75.93 78.5
72.3 77.70 71.6
72.3 77.74 71.6
72.3 77.77 71.5
Lastly, the hydrolytic profiles of Gen8 and Gen9 (Fig. 5) as refer-
ence compounds were compared with the time course of the
hydrolysis of pustulan (Fig. 4). Gen8 was hydrolyzed in the order
of DP 2 and DP 6 > DP 3 and DP 5 > DP 4, and Gen9 in the order
of DP 2 and DP 7 > DP 3 and DP 6 > DP 4 and DP 5 (Fig. 5). This indi-
cates that the major cleavages occur at bonds 2, 3, and 4 through a
reducing terminus. The cleavage pattern was substantially similar
72.5 78.8
63.6
Gen5 b-OMe CH3
60.3
60.3
1
2
3
4
5
106.2
105.7
105.7
75.89 78.53 72.3 77.7
75.89 78.45 72.3 77.7
75.89 78.45 72.3 77.7
71.6
71.6
71.6
71.5
63.6
105.84 75.89 78.45 72.3 77.7
105.77 75.94 78.53 72.5 78.8
ꢀ
to that of DP 10 derived from pustulan (Fig. 4D).
Gen6 b-OMe CH3
1
2
3
4
5
6
106.2
105.7
105.7
105.8
105.8
105.8
75.88 78.5
75.88 78.4
75.88 78.4
75.88 78.4
75.88 78.4
75.92 78.5
72.3 77.7
72.3 77.7
72.3 77.7
72.3 77.7
72.3 77.7
72.5 78.8
71.6
71.6
71.6
71.6
71.5
63.6
3. Discussion
The aim of the present work was to elucidate the hydrolytic
mode of action of purified b-(1?6)-glucanase on a typical linear
b-(1?6)-glucan pustulan. For this purpose, the enzyme was puri-
fied from the culture filtrate of P. multicolor. In this process, sub-
strate affinity chromatography using Gen4 as a ligand was shown
to be a powerful technique for the purification of b-(1?6)-glucan-
ase. The highly purified enzyme (244-fold purified, 164.1 U/mg)
gave a single protein band at 51 kDa by SDS–PAGE and was used
for analysis of its hydrolytic mode of action on pustulan. The
enzyme showed activity only against substrates with a b-(1?6)-
glycosidic linkage.
a
Numbers indicate the position of the glucose residues; the glucose residue that
is attached to the methyl group is represented as 1.
HPAEC-PAD. Pustulan was shown to have an average degree of
polymerization (DP) of 320, within the DP 200–550 range
(Fig. 4A). In the early stage (after 3 min) of incubation, the glucan
was quickly hydrolyzed to form a broad peak corresponding to
DP 45 in DP 15–100 range, which was designated a short-chain
ꢀ
ꢀ
Table 4
1H- and 13C-chemical shifts of p-nitrophenyl b-gentiooligosides (DP 2-6) in D2O solution (25 °C)
Compound
Residuea
Chemical shift (d)
b
C-1
C-2
C-3
C-4
C-5
C-6
o-ph
m-ph
p-ph
C–O
H-1
J1,2
Gen2 b-pNP
pNP
1
2
119.4
128.9
145.4
164.4
102.2
105.6
75.5
75.9
78.2
78.5
72.0
72.5
78.3
78.7
71.2
63.6
5.29
4.48
6.7
8.0
Gen3 b-pNP
Gen4 b-pNP
pNP
1
2
119.5
119.5
129.0
129.0
145.4
145.4
164.4
164.5
102.1
105.8
105.6
75.6
75.94
75.89
78.2
78.4
78.5
72.2
72.2
72.5
78.1
77.7
78.7
71.7
71.3
63.6
5.29
4.50
4.46
5.5
8.0
7.9
3
pNP
1
2
3
4
102.1
105.7
105.8
105.7
75.5
78.2
78.4
78.4
78.5
72.16
72.20
72.3
78.1
77.7
77.7
78.7
71.7
71.6
71.5
63.6
5.30
4.51
4.48
4.46
7.6
7.9
7.9
8.0
75.92
75.88
75.86
72.4
Gen5 b-pNP
Gen6 b-pNP
pNP
1
2
3
4
119.5
119.5
129.0
129.1
145.5
145.5
164.5
164.5
102.2
75.6
78.2
78.4
78.4
78.4
78.5
72.19
72.25
72.33
72.21
72.5
78.1
71.7
71.6
71.8
71.4
63.6
5.31
4.51
4.49
4.47
4.51
7.4
7.9
8.3
8.0
7.9
105.73
105.80
105.83
105.68
75.91
75.87
75.83
75.93
77.72
77.72
77.75
78.7
5
pNP
1
2
3
4
102.2
105.74
105.8
105.8
105.8
105.71
75.6
78.2
78.4
78.4
78.4
78.4
78.5
72.20
72.25
72.34
72.21
72.31
72.5
78.1
71.70
71.65
71.8
71.65
71.4
5.31
4.51
4.49
4.48
4.52
4.52
7.7
7.4
9.2
8.0
7.6
7.6
75.91
75.87
75.84
75.93
75.87
77.72
77.72
77.74
77.8
5
6
78.7
63.6
a
Numbers indicate the position of the glucose residues; the glucose residue that is attached to p-nitrophenyl group is represented as 1.
J1,2, coupling constants given in Hz.
b

T. Hattori et al. / Carbohydrate Research 366 (2013) 6–16
11
Table 6
Table 5
Frequency and relative activity of the b-(1?6)-glucanase-catalyzed hydrolysis of
gentiooligosaccharides
Frequency and relative activity of the b-(1?6)-glucanase-catalyzed hydrolysis of
methyl b-gentiooligosides and p-nitrophenyl b-gentiooligosides
Compound Frequency of hydrolysisa (%)
Relative activityb (%)
<1
Compound
Frequency of hydrolysisa (%)
Relative activityb (%)
4
Gen2
Gen2 b-OMe
100
100
G – G – CH3
G – G⁄
Gen3
15
21
Gen3 b-OMe
Gen4 b-OMe
Gen5 b-OMe
Gen6 b-OMe
Gen2 b-pNP
Gen3 b-pNP
Gen4 b-pNP
Gen5 b-pNP
Gen6 b-pNP
13
67
81
100
1
a
A
a + A = 100
(a < A)
9
91
G – G – G⁄
G – G – G – CH3
Gen4
Gen5
Gen6
68 32
G – G – G – G⁄
5
39 56
G – G – G – G – CH3
77
a
A
22
a + A = 78
(a < A)
31 30 39
G – G – G – G – G⁄
G – G – G – G – G – CH3
100
a
22
A
13
a + A = 65
12 31 30 27
G – G – G – G – G – G⁄ (a < A)
G – G – G – G – G – G – CH3
a
The frequency of the enzymatic cleavage of the indicated glycosidic linkages
100
G – G – pNP
was estimated by measuring the amount of lower gentiooligosaccharides liberated
from the corresponding gentiooligosaccharides.
The hydrolytic rate of Gen6 was arbitrarily set at 100. G, glucose residue; G⁄,
b
2
glucose residue with a reducing end.
39 61
G – G – G – pNP
22
39
100
90 10
to bonds 2 and 3.As mentioned above, our purpose was to elucidate
the mechanism of the hydrolysis of pustulan by b-(1?6)-glucanase
based on kinetic and product analyses. Enzymatic hydrolysis of
glycosidic bonds usually occurs with two possible stereochemical
outcomes namely, inversion or retention of the anomeric
configuration at the cleavage site.25 The hydrolytic mechanism of
the enzyme was first evaluated by 1H NMR. In the initial stage of
the reaction, the b-anomer was formed and was then easily con-
G – G – G – G – pNP
38 62
G – G – G – G – G – pNP
25 29 46
G – G – G – G – G – G – pNP
a
The frequency of the enzymatic cleavage of the indicated glycosidic linkages
verted into the
a-anomer by mutarotation. This indicates that
was estimated by measuring the amount of lower methyl b-gentiooligosides and p-
nitrophenyl gentiooligosides liberated from the corresponding methyl b-gentiool-
igosides and p-nitrophenyl b-gentiooligosaccharides, respectively.
The hydrolytic rate of Gen6 b-OMe and Gen6 b-pNP was arbitrarily set to 100. G,
glucose residue.
the hydrolysis of pustulan occurs with retention of the anomeric
configuration. Pitson et al.14 have reported that b-(1?6)-glucanase
from Acremonium persicinum acts with retention of the anomeric
center, releasing products in the b-configuration. Furthermore,
the pathway of the hydrolysis of pustulan by the purified enzyme
could be proposed on the basis of HPAEC-PAD analysis of the time
course of hydrolysate production (Scheme 2). The enzyme was
shown to achieve complete hydrolysis of the glucan in three steps
as follows. In the initial stage, the enzyme cleaved the glucan with
a pattern resembling that of an endo-hydrolase to quickly produce
b
The present method for obtaining a series of gentiooligosides
(DP 3-6) starting from pre-prepared gentiooligoside (DP 2) incor-
porated a simple one-pot reaction process and ensured the regiose-
lectivity of glycosylation, although it was not always sufficient. The
resulting synthetic methyl b-gentiooligosides and p-nitrophenyl
b-gentiooligosides were very useful as substrates for providing
analytical insight into how the enzyme actually cleaves pustulan.
From the cleavage frequencies, it was shown that methyl b-genti-
ooligosides (DP 2-6) and p-nitrophenyl b-gentiooligosides (DP 2-6)
were cleaved through the terminal glycoside and that the number
of the susceptible bond varied with the chain length of substrate.
However, the cleavage pattern of methyl b-gentiooligosides (DP
4-6) was somewhat different from that of p-nitrophenyl b-genti-
ooligosides (DP 4-6). For example, the cleavage of Gen5 b-OMe oc-
curred in almost equal proportion at bonds 2, 3, and 4, whereas
Gen5 b-pNP was preferentially cleaved at bonds 3 and 4, but not
cleaved at all at bond 2. This indicates that the presence of a
methyl group does not influence enzyme action, whereas that of
a p-nitrophenyl group shifts the cleavage position for one glucosyl
residue to the non-reducing end. Taking these data into account,
the cleavage frequencies of gentiooligosaccharides were proposed
on the basis of the assumption that the glucose residue at the
non-reducing terminus of gentiooligosaccharides with DP 4 or
higher is not hydrolyzed by the enzyme. The cleavage pattern of
gentiooligosaccharides is approximated to that of Genn b-OMe
rather than Genn b-pNP. Thus, cleavage at bond 1 gradually de-
creases with extension of the DP and the cleavage location shifts
ꢀ
a short-chain glucan (DP 45) as an intermediate after 3 min of
incubation. In the midterm stage, the resulting short-chain glucan
was further cleaved into two peak fractions corresponding to DP
15-10 and DP 2-4 within 30 min. During this process, it can be
envisaged how the enzyme initially attacks the glucan so quickly.
Notably, the relative hydrolytic activity of the enzyme toward
Gen8 was 3, whereas that toward pustulan was 100, representing
a 33-fold difference. Once the chain length of DP 15-10 was
reached, the preferred cleavage occurred at bonds 2, 3, and 4 from
the reducing terminus of the chain to afford DP 2-4 with great
regularity. Only a low level of glucose was detected despite the
subsequent reaction. In the midterm stage, the cleavage pattern
ꢀ
of DP 10 derived from pustulan (Fig. 4D) was substantially similar
to those of Gen8 and Gen9 (Fig. 5). In the final stage of the reaction,
the lower oligomers (DP 3 and DP 4) underwent hydrolysis extre-
mely slowly to produce glucose and Gen2 as the main products, be-
cause of the lower hydrolytic activity of the enzyme on Gen3 and
Gen4. It took prolonged incubation to reach this stage (Fig. 4F). In
this way, the enzyme varies its mode of action depending on the
chain length derived from the glucan during the whole course of
the reaction. Montero et al.20 demonstrated that a b-(1?6)-glucan-
ase from Trichoderma harzianum, namely BGN16.3, has an

12
T. Hattori et al. / Carbohydrate Research 366 (2013) 6–16
the reaction. Longer polymers are initially hydrolyzed by an
endo-hydrolytic mode of action. In the subsequent reaction, the
resulting short-chain glucan is hydrolyzed by a pattern resembling
a saccharifying-hydrolase to produce DP 2-4 with great regularity;
ultimately, DP 2-4 are cut up to produce glucose and Gen2. As a
result, it has been revealed that the hydrolytic cooperation of both
an endo-type and a saccharifying-type reaction of a single enzyme,
which plays a bifunctional role, leads to complete hydrolysis of the
b-(1?6)-glucan pustulan.
4. Experimental
4.1. Materials
A crude enzyme preparation from P. multicolor IAM7153 (Insti-
tute of Molecular and Cellular Biosciences, The University of
Tokyo) was kindly supplied by Amano Enzyme, Inc. (Gifu, Japan).
Cellulase from H. jecorina was purchased from NagaseChemteX
Co. (Osaka, Japan). Gentiobiose, methyl b-
D-glucopyranoside,
p-nitrophenyl b- -glucopyranoside, glucan from Saccharomyces
D
cerevisiae, carboxymethylcellulose, and curdlan from Alcaligenes
faecalis were purchased from Sigma Chemical Co. (St. Louis, MO,
USA). Pustulan was purchased from Calbiochem (San Diego, CA,
USA) and prepared for the removal of insoluble material as follows:
the pustulan was dissolved in water (10 mg/mL) and boiled for
10 min. After cooling, the resulting precipitate was removed by
centrifugation (5000g, 15 min). The supernatant was then lyophi-
lized and used as substrate for the b-(1?6)-glucanase assay. The
lyophilized pustulan was also fractionated by 50% ethanol precip-
itation to afford a water-soluble pustulan for hydrolytic analysis.
All other commonly used chemicals were obtained from commer-
cial sources.
Figure 3. 1H NMR spectra of the time course of pustulan hydrolysis by the purified
b-(1?6)-glucanase. (A) Initial substrate, (B) 20 min, (C) 1 h, (D) 4 h, and (E) 20 h
incubation after addition of the enzyme.
a and b indicate signals from H-1a and H-
1b protons, respectively, of gentiooligosaccharides produced by pustulan hydrolysis
by the enzyme. The enzyme reaction was performed in a 3-mm sample tube at
25 °C. Pustulan was dissolved in D2O (0.65%, w/v) and then the enzyme solution
(10 lL, 72 mU) was added.
4.2. Analytical methods
endo-hydrolytic mode of action by HPLC analysis of the products
from pustulan.
In conclusion, b-(1?6)-glucanase from P. multicolor was puri-
fied and the hydrolytic manner of the enzyme toward b-(1?6)-
glucan was elucidated in real time during the whole course of
MALDI-TOF mass spectra were obtained on an AutoFlex instru-
ment (Bruker Daltonics, Bremen, Germany). The spectra were mea-
sured in positive linear mode with 2,5-dihydroxybenzoic acid
(10 mg/mL) in H2O/ethanol (70:30) as the matrix solution. The
Figure 4. HPAEC-PAD analysis of the hydrolytic products of pustulan obtained with the purified b-(1?6)-glucanase. (A) Initial substrate, (B) 3 min, (C) 10 min, (D) 30 min, (E)
2 h, and (F) 96 h after incubation after the addition of enzyme. Numbers indicate the DP of gentiooligosaccharides.

T. Hattori et al. / Carbohydrate Research 366 (2013) 6–16
13
8
8
0 min
1.5 min
G-G-G-G-G-G-G-G*
6
2
7
5
3 4
1
0 min
1.5 min
9
9
G-G-G-G-G-G-G-G-G*
2
7
6
3
4
8
5
Cleavage site
G Glc
1
G* Glc (reducing end)
− β-(1 6)-linkage
Retention time (min)
Retention time (min)
Figure 5. HPAEC-PAD analysis of the hydrolytic products of Gen8 (top) and Gen9 (bottom) obtained with the purified b-(1?6)-glucanase. Left, initial substrate; right, 1.5 min
after the addition of enzyme. Numbers indicate the DP of gentiooligosaccharides.
β-(1 6)-glucan
G-G-G·····G-G-G-G-G-G-G-G-G·····G-G-G*
Fast
DP>15
DP 15
Endo-type
G-G-G-G-G-G-G-G-G-G-G-G-G-G-G*
Slow
G-G-G-G-G-G-G-G-G-G* DP 10
Very slow
Saccharifyng-type
DP=3, 4
G-G-G-G* G-G-G*
Extremely slow
Cleavage site
G
Glc
G* Glc (reducing-end)
β-(1 6)-linkage
−
G-G* G* DP=1, 2
Scheme 2. Proposed mechanism of the hydrolytic action of b-(1?6)-glucanase from P. multicolor on pustulan.
sample solutions were mixed with the matrix solution (1:4 v/v),
and then a 1- L droplet was applied to a stainless target plate
v) to 60% (v/v), and eluent C was kept at 40% (v/v); 50-60 min, 100%
l
(v/v) eluent B; 60–70 min, 60% (v/v) eluent A, and 40% (v/v) eluent
C. In both conditions, eluents A, B, and C were deionized water,
0.1 M NaOH containing 1 M sodium acetate, and 0.2 M NaOH,
respectively. The flow rate was 1.0 mL/min. 1H and 13C NMR spec-
tra of each sample in D2O were recorded on a JEOL JNM-LA 500
spectrometer. Chemical shifts (d) were expressed relative to
sodium 3-(trimethylsilyl)-propionate (TPS) as an external stan-
dard. ESI-MS spectra were recorded on a JMS-T100LC mass spec-
trometer. The samples were injected in methanol (0.2–0.5 mg/
mL) directly into the spectrometer.
and dried at room temperature. A mass calibration procedure
was carried out before the analysis of samples by using protein cal-
ibration standard II (Bruker Daltonics). The HPLC analysis was car-
ried out on an Asahipak GS-220 HQ column (7.5 Â 300 mm, Japan)
with a Jasco Intelligent system liquid chromatograph and detection
at 300 nm. The bound material was eluted with H2O at a flow rate
of 0.9 mL/min at 40 °C. HPAEC-PAD analysis was carried out using
a DIONEX DX-500 system (Dionex, Sunnyvale, CA) fitted with a
PAD (ED-40) and a CarboPac PA1 column (4.0 Â 250 mm). Bound
material was eluted by the following gradients. Condition 1:
0–40 min, the ratio of eluent A was decreased from 60% (v/v) to
45% (v/v), the ratio of eluent B was increased from 0% (v/v) to
15% (v/v), and eluent C was kept at 40% (v/v); 40–50 min, 100%
(v/v) eluent B; 50–60 min, 60% (v/v) eluent A, and 40% (v/v) eluent
C. Condition 2: 0–50 min, the ratio of eluent A was decreased from
60% (v/v) to 0% (v/v), the ratio of eluent B was increased from 0% (v/
4.3. Enzyme assays
b-(1?6)-Glucanase was assayed as follows. A mixture contain-
ing 180
acetate buffer (pH 5.5) and 20
enzyme was incubated for 15 min at 40 °C. The reducing sugars
lL of substrate (5 mg/mL of pustulan) in 50 mM sodium
l
L of an appropriate amount of

14
T. Hattori et al. / Carbohydrate Research 366 (2013) 6–16
produced were determined directly by Somogyi-Nelson’s meth-
od.26,27 One unit was determined as the amount of enzyme catalyz-
the reaction mixture. The pH stability was determined by incubat-
ing the enzyme at 4 °C for 24 h at each pH, and then measuring the
activity under the standard conditions. The buffers used were
Tris–HCl (pH 2.0–3.0), sodium acetate (pH 3.0–6.0), and sodium
phosphate (pH 7.0–9.0). The optimum temperature of the activity
was measured under the standard conditions at various tempera-
tures (4–80 °C). Thermostability was determined by incubating
the enzyme at each temperature for 1 h in 50 mM sodium acetate
buffer (pH 5.5), and then measuring the activity under the standard
conditions.
ing the formation of 1
substrate under the conditions of the assay. b-Glucosidase was
measured as follows: 100 L of 10 mM p-nitrophenyl b- -glucoside
solution, 250 L of 100 mM sodium acetate buffer (pH 4.0), and
125 L of H2O were premixed. Then, enzyme solution (25 L)
lmol of D-glucose per min from the
l
D
l
l
l
was added to the solution, which was kept at 40 °C for 15 min.
One-tenth of the reaction mixture was removed at 5-min intervals
and immediately transferred to a microplate containing 50 lL of
1 M Na2CO3 to stop the reaction, and the p-nitrophenol liberated
was determined spectrophotometrically at 420 nm. One unit of
activity was defined as the amount of enzyme releasing 1 lmol
4.7. Enzymatic synthesis of methyl b-gentiooligosides (DP 2-6,
Gen2-6 b-OMe)
of p-nitrophenol per min.
Glcb-OMe (15 g, 77.2 mmol) was dissolved in 21.2 mL of
100 mM sodium acetate buffer (pH 4.0) and b-glucosidase
(17.4 mL, 776 U) in crude cellulase from H. jecorina was added.
The solution was incubated without stirring for 23 h at 40 °C, and
the reaction was terminated by boiling for 5 min. The reaction
mixture was applied to a charcoal-Celite column (4.5 Â 50 cm)
equilibrated with water. The column was washed with 1.5 L of
water and eluted with a linear gradient of 0% (3 L) to 7.5% (3 L) eth-
anol (flow rate 7 mL/min; a fraction size 50 mL/tube). Elution was
monitored at 485 nm, with carbohydrate content determined by
the phenol–H2SO4 method. The eluate (550–4115 mL) was concen-
trated and applied to charcoal-Celite chromatography again. The
eluate (850–1550 mL) was concentrated and lyophilized, and affor-
ded 770 mg of Gen2 b-OMe28 in 5.6% yield based on the Glcb-OMe
added. As an acceptor and a donor, the synthetic Gen2 b-OMe
4.4. Preparation of Gen4-coupled gel
Gen4-coupled Toyopearl AF-Amino-650M was prepared as fol-
lows: Gen4 (580 mg in 2 mL of 0.2 M K2HPO4) was added to pre-
swollen gel (1 g) and then NaCNBH3 (200 mg) was added. The gel
in the solution was incubated at 60 °C overnight with gentle agita-
tion, and then washed with water (20 mL) three times and sodium
borate buffer (20 mL, pH 8.2) three times. For acetylation of the
free amino group, 0.2 M sodium acetate (8 mL) and acetic anhy-
dride (4 mL) were added to the gel and allowed to react for
30 min at 0 °C. Next, acetic anhydride (4 mL) was added to the
solution containing the gel, which was then kept for 30 min at
room temperature with gentle agitation. The gel was washed suc-
cessively with H2O, 0.1 M NaOH, and H2O (three times for each)
and packed into a column (1 mL).
(627 mg, 1.8 mmol) was dissolved in 340
lL of 50 mM sodium ace-
tate buffer (pH 4.0) and purified b-(1?6)-glucanase (100
lL,
4.5. Purification of b-(1?6)-glucanase
28.3 mU) was added. After the solution was incubated for 60 h at
50 °C, the reaction mixture was boiled for 15 min. To remove
reducing sugar in the reaction mixture by chromatography, pyr-
idylamination was performed as follows: Solution A was prepared
Unless otherwise indicated, all steps were carried out at 4 °C. The
P. multicolor crude enzyme (720 mg) was dissolved in 50 mM so-
dium acetate buffer (pH 5.5), brought to 60% saturation with solid
ammonium sulfate, and left to stand for 30 min. The precipitate
was removed by centrifugation (5000 g, 10 min). The supernatant
was brought to 80% saturation with solid ammonium sulfate and
left to stand for 30 min. The precipitate was dissolved in 50 mM
sodium acetate buffer (10 mL, pH 5.5), dialyzed against H2O, and
lyophilized. The enzyme powder (80 mg) was dissolved in 20 mM
sodium acetate buffer (1 mL, pH 4.0), applied to a CM-Sepharose
Fast Flow column (2.5 Â 30 cm) equilibrated with 20 mM sodium
acetate buffer (pH 4.0), and eluted with the same buffer (200 mL)
at a flow rate of 1 mL/min. The eluent was collected in 10-mL frac-
tions. b-(1?6)-Glucanase activity was measured by a standard as-
say using each fraction as an enzyme solution. Fractions showing
b-(1?6)-glucanase activity (tubes 6–10) were pooled and concen-
trated with an Amicon PM-30 membrane. The concentrated
enzyme fraction was applied to a Gen4-coupled affinity column
(1 mL) equilibrated with 20 mM sodium acetate buffer (pH 5.5)
containing 45 mM NaCl. After the column was washed with 20 mL
of equilibration buffer, the adsorbed material was eluted with a lin-
ear gradient from 45 to 800 mM NaCl in the same buffer over a vol-
ume of 20 mL, and then with 800 mM NaCl in the same buffer at a
flow rate of 0.1 mL/min. The eluent was collected in 0.5-mL frac-
tions. Fractions showing b-(1?6)-glucanase activity (tubes 52–
54) were combined and concentrated with a Nanosep 10k Omega
membrane. The enzyme purity was checked by SDS–PAGE.
by dissolving 0.5 g of 2-aminopyridine in 380
tion B was prepared by dissolving 40 mg of NaBH3CN in Solution A
and adding 100 L of H2O. Solution A was added to the reaction
mixture, kept at 100 °C for 15 min, and then 295 L of solution B
lL of 35% HCl. Solu-
l
l
was added. The mixture was incubated for 24 h at 70ꢁC. After pyr-
idylamination, the mixture was loaded onto a TOYOPEARL HW-
40S column (4.5 Â 90 cm) equilibrated with 25% methanol at a
flow rate of 1.5 mL/min. The eluate (1125–2475 mL) was collected,
evaporated to remove methanol, and loaded onto a charcoal-Celite
column (4.5 Â 90 cm) equilibrated with H2O. The column was
washed with 1.5 L of water and then eluted with a linear gradient
of 5% (5 L) to 35% (5 L) ethanol (flow rate, 7 mL/min; fraction size,
50 mL/tube). The eluate showed six main peaks, MG1 (tubes
18–26), MG2 (tubes 62–73), MG3 (tubes 95–103), MG4 (tubes
108–115), MG5 (tubes 120–125), and MG6 (tubes 129–134), which
were concentrated and lyophilized. Fraction MG1 was recovered as
Glc b-OMe (142.3 mg) and fraction MG2 was recovered as Gen2
b-OMe (242.2 mg, 38.6% recovery). Fractions MG3, MG4, MG5, and
MG6 gave Gen3 b-OMe (63.4 mg, 7.0%), Gen4 b-OMe (51.5 mg,
4.3%), Gen5 b-OMe (22.9 mg, 1.5%), and Gen6 b-OMe (15.0 mg,
0.8%), respectively.
The structures of synthetic Gen2–Gen6 b-OMe were evaluated
by 1H and 13C NMR analyses in D2O. Characteristic signals were
commonly observed in the region 4.20–4.55 ppm. H-1b signals of
the products were observed at 4.39–4.42 ppm (d, J1,2 = 8 Hz). The
peaks of internal anomeric protons were observed as overlapping
signals at 4.55–4.51 ppm (d, J1,2 = 8 Hz), characteristic of the b-
(1?6)-linkage from the corresponding methyl b-gentiooligosides.
Furthermore, overlapping signals at 4.21–4.25 ppm were observed.
The signals were indicative of low-field shifts of internal H-6
protons. In addition, peaks of methyl group were observed at
4.6. Characterization of b-(1?6)-glucanase
The effect of pH on the activity of the purified b-(1?6)-glucan-
ase was measured under the standard conditions of pustulan
hydrolysis while changing the pH of the 50 mM buffers used in

T. Hattori et al. / Carbohydrate Research 366 (2013) 6–16
15
3.58–3.60 ppm (s, 3H). The structures of these methyl b-gentiooli-
gosides were further confirmed by 13C NMR analysis, as shown in
Table 3. Each peak could be assigned to the corresponding carbon
atom of a methyl b-oligoside with a b-(1?6) linkage. No signals
derived from other linkages were detected. In addition, ESI-MS
analysis of Gen2 b-OMe, Gen3 b-OMe, Gen4 b-OMe, Gen5 b-OMe,
and Gen6 b-OMe showed molecular ions at m/z 379.12125 (Calcd
by H2O to match that of the pustulan solution (5 mg/mL). The
hydrolysate and enzyme activity on these substrates was deter-
mined by measuring the amount of reducing sugar by the Somo-
gyi-Nelson method.26,27 Methyl b-gentiooligosides (DP 2-6),
p-nitrophenyl b-gentiooligosides (DP 2-6), and gentioligosaccha-
rides (DP 2-6) were used to analyze the frequency of the enzymatic
cleavage of the b-(1?6)-linkage and relative hydrolytic rate. Each
for C13H24NaO11, 379.12163), 541.17498 (Calcd for C19H34NaO16
,
substrate was dissolved in H2O (25
25 mM sodium acetate buffer (200
the enzyme solution (25 L, 59 mU). The reaction was conducted
l
L, 10 mM) and mixed with
541.17445), 703.22668 (Calcd for 25H44NaO21 703.22728),
C
,
lL, pH4.0) and then added to
865.28050 (Calcd for C31H54NaO26, 865.28010), and 1027.33524
(Calcd for C37H64NaO31, 1027.33292), respectively, arising from
the [M+Na]+ ions.
l
at 40ꢁC. The amount of each product formed at an early stage
(within 27% hydrolysis) from the initial substrate during incuba-
tion with the enzyme was analyzed by HPAEC-PAD (Gen2-6
b-OMe and Gen2-6, condition 1) and HPLC (Gen2-6 b-pNP). From
the peak areas on the chromatogram of the digest of each substrate
with enzyme, the amount of the products was calculated. The
amount of each product increased linearly with time in the initial
stage of the reaction. On the basis of these data, the frequency and
relative hydrolytic rate of the b-(1?6)-glucanase-catalyzed cleav-
age of glycosidic linkages were determined.
4.8. Enzymatic synthesis of p-nitrophenyl b-gentiooligosides
(DP 3-6, Gen3-6 b-pNP)
Glc b-pNP (1356 mg, 4.5 mmol) was dissolved in 33.7 mL of
83 mM sodium acetate buffer (pH 4.0) and b-glucosidase
(11.3 mL, 24.9 U) in crude cellulase from H. jecorina was added.
The mixture was incubated without stirring for 6 h at 40 °C. The
reaction was terminated by boiling for 15 min, and the solution
was loaded onto a Toyopearl HW-40S column (6 Â 100 cm) equil-
ibrated with 25% (v/v) methanol at a flow rate of 1.5 mL/min. The
eluate was collected in 25-mL fractions and monitored by measur-
ing the absorbance at 300 nm (p-nitrophenyl group). Peak fractions
(3150–3400 mL) containing the target product were pooled, con-
centrated, and lyophilized. Gen2 b-pNP29 (158.1 mg) was obtained
in 15.2% yield based on the initial substrate. The synthetic Gen2 b-
pNP (278 mg, 0.6 mmol) and Gen3 (757 mg, 1.5 mmol) were dis-
solved in 2.8 mL of 55 mM sodium acetate buffer (pH 5.5) and P.
multicolor b-1,6-glucanase solution (0.2 mL, 2.4 U) was added.
The mixture was incubated for 116 h at 4 °C. The reaction was
terminated by boiling for 15 min and loaded onto a Toyopearl
HW-40S column (6 Â 100 cm) equilibrated with 25% methanol at
a flow rate of 1.5 mL/min (25 mL/tube). The chromatogram showed
five peaks eluted in the following order: PG6 (1700–1800 mL), PG5
(1850–1950 mL), PG4 (2075–2200 mL), PG3 (2450–2550 mL), and
PG2 (3300–3525 mL). Fractions PG4, PG3, and PG2 were individually
concentrated and lyophilized to afford Gen4 b-pNP (66.9 mg), Gen3
b-pNP (7.7 mg), and Gen2 b-pNP (159.7 mg), respectively, in 14.2%,
2.1%, and 57.4% yields based on the acceptor. PG6 and PG5 were
separately concentrated and loaded onto a reverse-phase ODS
column (1.5 Â 3.1 cm) equilibrated with H2O. After washing the
column with H2O (30 mL), the adsorbed glucoside was eluted with
methanol (30 mL), and the eluate was collected in 2-mL fractions
and monitored by measuring the absorbance at 300 nm (p-nitro-
phenyl group). Peak fractions (32-50 mL) were individually
combined, concentrated, and lyophilized to afford Gen6 b-pNP
(7.8 mg) and Gen5 b-pNP (14.9 mg) in 1.7% and 2.6% yields, respec-
tively, based on the acceptor.
The pattern of pustulan hydrolysis by the enzyme was analyzed
as follows: purified pustulan was dissolved in H2O (900
lL, 5 mg/
mL) and then added to enzyme solution (100 L, 53 mU) in
l
50 mM sodium acetate buffer (pH 5.5). The reaction was performed
at 40ꢁC and monitored by HPAEC-PAD analysis (condition 2). The
hydrolytic profile of Gen8 and Gen9 was analyzed as follows: each
substrate was dissolved in H2O (25
sodium acetate buffer (200 L, pH 4.0) and then added to enzyme
solution (25 L, 59 mU). The reaction was conducted at 40ꢁC. The
lL, 10 mM), mixed with 25 mM
l
l
amount of each product formed at an early stage (Gen8; 10% hydro-
lysis, Gen9; 35% hydrolysis) was analyzed by HPAEC-PAD (condi-
tion 1)
DP of pustulan was estimated as follows: purified pustulan was
hydrolyzed by b-(1?6)-glucanase and the hydrolysate was ana-
lyzed by HPAEC-PAD (condition 2). Standard curve was generated
from the retention time of clearly resolved peaks and correspond-
ing DPs that have less than 56. Higher DPs were calculated from
the extrapolated standard curve.
Acknowledgments
We thank Amano Enzyme, Inc. (Gifu, Japan) for the gift of the
enzyme preparation from P. multicolor. This work was supported
by a research grant (Agribusiness) from the Ministry of Agriculture,
Forestry, and Fisheries of Japan.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.carres.2012.11.
002.
The structures of synthetic Gen2-Gen6 b-pNP were evaluated by
1H NMR and 13C NMR as in Table 4. ESI-MS analysis of Gen2 b-pNP,
Gen3 b-pNP, Gen4 b-pNP, Gen5 b-pNP, and Gen6 b-pNP showed
References
molecular ions at m/z 486.12527 (Calcd for
486.12236), 648.17548 (Calcd for 24H35NNaO18
810.22788 (Calcd for C30H45NNaO23, 810.22800), 972.28091 (Calcd
for 36H55NNaO28 972.28083), and 1134.33385 (Calcd for
42H65NNaO33 1134.33365), respectively, arising from the
[M+Na]+ ions.
C
18H25NNaO13
,
C
, 648.17518),
1. Kapteyn, J. C.; Montijn, R. C.; Vink, E.; De la Cruz, J.; Llobell, A.; Douwes, J. E.;
Shimoi, H.; Lipke, P. N.; Klis, F. M. Glycobiology 1996, 6, 337–345.
2. Lindberg, B.; McPherson, J. Acta Chem. Scand. 1954, 8, 985–988.
3. Henrissat, B.; Bairoch, A. Biochem. J. 1993, 293, 781–788.
4. Davies, G.; Henrissat, B. Structure 1995, 3, 853–859.
5. Henrissat, B.; Bairoch, A. Biochem. J. 1996, 316, 695–696.
6. Henrissat, B.; Davies, G. J. Curr. Opin. Struct. Biol. 1997, 7, 637–644.
7. Reese, E. T.; Parrish, F. W.; Mandels, M. Can. J. Microbiol. 1962, 8, 327–334.
8. Shibata, Y.; Fukimbara, T. J. Ferment. Technol. 1973, 51, 216–226.
9. Yamamoto, S.; Kobayashi, R.; Nagasaki, S. Agric. Biol. Chem. 1974, 38, 1493–
1500.
C
,
C
,
4.9. Action of b-(1?6)-glucanase
The hydrolytic activity of purified b-(1?6)-glucanase was eval-
uated by using several glucans and an oligosaccharide (Gen8) un-
der standard conditions. Before the assay, Gen8 was dissolved in
H2O and then the reducing power in the solution was adjusted
10. Santos, T.; Villanueva, J. R.; Nombela, C. J. Bacteriol. 1977, 129, 52–58.
11. Katohda, S.; Suguki, F.; Katsuki, S.; Sato, T. Agric. Biol. Chem. 1979, 43, 2029–
2034.
12. Hiura, N.; Nakajima, T.; Matsuda, K. Agric. Biol. Chem. 1987, 51, 3315–3321.

16
T. Hattori et al. / Carbohydrate Research 366 (2013) 6–16
13. Lora, J. M.; De la Cruz, J.; Llobell, A.; Benítez, T.; Pintor-Toro, J. A. Mol. Gen.
Genet. 1995, 247, 639–645.
21. Himmel, M. E.; Ding, S. Y.; Johnson, D. K.; Adney, W. S.; Nimlos, M. R.; Brady, J.
W.; Foust, T. D. Science 2007, 315, 804–807.
14. Pitson, S. M.; Seviour, R. J.; McDougall, B. M.; Stone, B. A.; Sadek, M. Biochem. J.
1996, 316, 841–846.
22. Vaaje-Kolstad, G.; Westereng, B.; Horn, S. J.; Liu, Z.; Zhai, H.; Sørlie, M.; Eijsink,
V. G. H. Science 2010, 330, 219–222.
15. Martin, K. L.; Unkles, S. E.; McDougall, B. M.; Seviour, R. J. Enzyme Microb.
Technol. 2006, 38, 351–357.
23. Reeve, A. In Starch Hydrolysis Products; Schenck, F. W., Hebeda, R., Eds.; VCH
Publishers: New York, 1992; p 79.
16. Montero, M.; Sanz, L.; Rey, M.; Llobell, A.; Monte, E. J. Appl. Microbiol. 2007, 103,
1291–1300.
24. Fujimoto, Y.; Hattori, T.; Uno, S.; Murata, T.; Usui, T. Carbohydr. Res. 2009, 344,
972–978.
17. Konno, N.; Sakamoto, Y. Appl. Microbiol. Biotechnol. 2011, 91, 1365–1373.
18. De La Cruz, J.; Pintor-Toro, J. A.; Benítez, T.; Llobell, A. J. Bacteriol. 1995, 177,
1864–1871.
25. McCarter, J. D.; Withers, G. S. Curr. Opin. Struct. Biol. 1994, 4, 557–564.
26. Somogyi, M. J. Biol. Chem. 1952, 195, 19–23.
27. Nelson, N. J. Methods Enzymol. 1955, 3, 85–86.
19. De La Cruz, J.; Llobell, A. Eur. J. Biochem. 1999, 265, 145–151.
20. Montero, M.; Sanz, L.; Rey, M.; Monte, E.; Llobell, A. FEBS J. 2005, 272, 3441–
3448.
28. Jansson, P. E.; Kenne, L.; Kolare, I. Carbohydr. Res. 1994, 257, 163–174.
29. Hrmova, M.; MacGregor, A. E.; Biely, P.; Stewart, R. J.; Fincher, G. B. J. Biol. Chem.
1998, 273, 11134–11143.