Protein engineering of Sulfolobus solfataricus maltooligosyltrehalose synthase to alter its selectivity

Article abstract of DOI:10.1021/jf0701279

Maltooligosyltrehalose synthase (MTSase) is one of the key enzymes involved in trehalose production from starch and catalyzes an intramolecular transglycosylation reaction by converting the α-1,4- to α,α-1,1-glucosidic linkage. Mutations at residues F206, F207, and F405 were constructed to change the selectivity of the enzyme because the changes in selectivity could reduce the side hydrolysis reaction of releasing glucose and thus increase trehalose production from starch. As compared with wild-type MTSase, F405Y and F405M MTSases had decreased ratios of the initial rate of glucose formation to that of trehalose formation in starch digestion at 75°C when wild-type and mutant MTSases were, respectively, used with isoamylase and maltooligosyltrehalose trehalohydrolase (MTHase). The highest trehalose yield from starch digestion was by the mutant MTSase having the lowest initial rate of glucose formation to trehalose formation, and this predicted high trehalose yield better than the ratio of catalytic efficiency for hydrolysis to that for transglycosylation.

Full text of DOI:10.1021/jf0701279

5588 J. Agric. Food Chem. 2007, 55, 5588−5594
Protein Engineering of Sulfolobus solfataricus
Maltooligosyltrehalose Synthase To Alter Its Selectivity
TSUEI-YUN FANG,*,† WEN-CHI TSENG,‡ CHING-HSING PAN,† YAO-TE CHUN,† AND
†
MEI-YING WANG
Department of Food Science, National Taiwan Ocean University, Keelung, Taiwan, and Department of
Chemical Engineering, National Taiwan University of Science and Technology, Taipei, Taiwan
Maltooligosyltrehalose synthase (MTSase) is one of the key enzymes involved in trehalose production
from starch and catalyzes an intramolecular transglycosylation reaction by converting the R-1,4- to
R,R-1,1-glucosidic linkage. Mutations at residues F206, F207, and F405 were constructed to change
the selectivity of the enzyme because the changes in selectivity could reduce the side hydrolysis
reaction of releasing glucose and thus increase trehalose production from starch. As compared with
wild-type MTSase, F405Y and F405M MTSases had decreased ratios of the initial rate of glucose
formation to that of trehalose formation in starch digestion at 75 °C when wild-type and mutant
MTSases were, respectively, used with isoamylase and maltooligosyltrehalose trehalohydrolase
(MTHase). The highest trehalose yield from starch digestion was by the mutant MTSase having the
lowest initial rate of glucose formation to trehalose formation, and this predicted high trehalose yield
better than the ratio of catalytic efficiency for hydrolysis to that for transglycosylation.
KEYWORDS: Maltooligosyltrehalose synthase; selectivity; site-directed mutagenesis; trehalose; substrate
specificity; Sulfolobus
INTRODUCTION
trehalose from starch, the yield of trehalose was limited to below
82% due to the side hydrolysis reactions catalyzed by MTSase
and MTHase (2-4). This suggests that engineering the substrate
specificities of both MTSase and MTHase to decrease the side
hydrolysis reactions would increase the trehalose yield.
Maltooligosyltrehalose synthase (EC 5.4.99.15, MTSase, also
known as glycosyltrehalose synthase, glycosyltrehalose-produc-
ing enzyme, and trehalosyl dextrin-forming enzyme) catalyzes
an intramolecular transglycosylation reaction to produce a
nonreducing maltooligosyltrehalose by converting the R-1,4-
glucosidic linkage at the reducing end of maltooligosaccharide
to an R,R-1,1-glucosidic linkage. Maltooligosyltrehalose can be
further hydrolyzed by maltooligosyltrehalose trehalohydrolase
(EC 3.2.1.141, MTHase, also known as glycosyltrehalose
trehalohydrolase, glycosyltrehalose-hydrolyzing enzyme, and
trehalose-forming enzyme) to produce trehalose by cleaving the
R-1,4-glucosidic linkage next to the R,R-1,1-linked terminal
disaccharide. However, both MTSase and MTHase catalyze a
side reaction of hydrolyzing maltooligosaccharides to release
glucose (G1).
Trehalose (R-D-glucopyranosyl-R-D-glucopyranoside) is a
nonreducing sugar that consists of an R,R-1,1-linkage between
two G1 molecules. The sugar protects proteins and cell
membranes from dry, heat, and osmotic changes and is gaining
more applications in many different areas, such as a sweetener
component, preservative, or stabilizer for food, cosmetics,
vaccines, and medicines (1). When thermophilic MTSase and
MTHase were combined with a debranching enzyme to produce
Previous kinetic studies have indicated that MTSase has 10
glucosyl binding subsites, numbered from +1 to -9, with the
catalytic site that is located between subsites +1 and -1 (5).
The MTSase catalytic mechanism involves three carboxyl
groups, Asp228, Glu255, and Asp443, in Sulfolobus acidocal-
daricus MTSase numbering (6). The intramolecular transgly-
cosylation catalyzed by MTSase is considered to consist of two
stages: (i) the cleavage of the R-1,4-glucosidic linkage and (ii)
the reconnection of forming the R,R-1,1-glucosidic linkage (6).
Asp443 first twists and deforms the G1 ring located at the subsite
-1 when the substrate is bounded to the active site. Glu255
and Asp228 may then function as the catalytic acid and base,
respectively, to cleave the glucosidic bond between subsites -1
and +1. The cleaved G1 unit was rotated in subsite +1 and
then reconnected to form an R,R-1,1-glucosidic linkage (6). The
hypothetical binding mode of carbohydrates in the active site
pocket of MTSase, which was proposed by Kobayashi et al.
(Figure 1) (6), has suggested that several amino acid residues
located near subsites +1 and -1 of the active site may affect
the rotation of the cleaved G1 unit prior to the reconnection to
form an R,R-1,1-linkage.
* To whom correspondence should be addressed. Tel: 886-2-2462-2192
ext. 5141. Fax: 886-2-2462-2586. E-mail: tyfang@mail.ntou.edu.tw.
† National Taiwan Ocean University.
The MTSase from Sulfolobus solfataricus ATCC 35092, also
known as P2, has been purified, characterized, and mutated in
‡ National Taiwan University of Science and Technology.
10.1021/jf0701279 CCC: $37.00
© 2007 American Chemical Society
Published on Web 06/14/2007

Altered Selectivity of Maltooligosyltrehalose Synthase
J. Agric. Food Chem., Vol. 55, No. 14, 2007 5589
Table 1. Nucleotide Sequences of Primers Used in Site-Directed
Mutagenesisa
mutation
nucleotide sequence of the primer (5′ f 3′)
F206W
F206Y
F207Y
F405M
F405S
F405W
CAAATTATAGGAGATGGTTCGCGGTAAATG
CAAATTATAGGAGATACTTCGCGGTAAATG
CAAATTATAGGAGATTCTACGCGGTAAATGATTTG
CAGCAATCATGGCTAAGGGCTATG
CATGCCAGCAATCTCCGCTAAGGGCTATG
CAGCAATCTGGGCTAAGGGCTATG
a Only the sequences of forward primers are shown. The sequence of reverse
primer for each mutation is complementary to its forward primer. Nucleotides for
the designed MTSase mutations are shown in bold. The silent mutation, decreasing
the melting temperature of the primer hairpin, is underlined.
E. coli Rosetta (DE3) at 37 °C in terrific broth medium supplemented
with 100 µg/mL ampicillin plus 34 µg/mL chloramphenicol as
previously described (7). Wild-type and mutant MTSases were purified
after cell disruption by using heat treatment, streptomycin sulfate
precipitation, Q-Sepharose anion exchange column chromatography,
and Sephacryl S-200 HR column chromatography as previously
described (8). The protein concentration was quantitated by Bradford’s
method (9) with BSA as standards.
Enzyme Kinetics. The values of kcat and KM for hydrolysis of G3
and transglycosylation of G4-G7 were determined at 60 °C in 50 mM
citrate-phosphate buffer at pH 5 by using 8-10 substrate concentrations
ranging from 0.1 KM to 8 KM as previously described (8). Values of
Figure 1. Schematic representation of the proposed binding model
between the S. acidocaldaricus MTSase and a substrate with an
R-1,4-
linkage (6, 8). The subsites are numbered according to the general subsite-
labeling scheme proposed for glycosyl hydrolases (24), in which the
substrate reducing end is at position
+1. The corresponding residues in
S. solfataricus ATCC 35092 are shown in parentheses.
our previous study (7, 8). Our previous results suggested that
the decreases in the ratios of hydrolysis to transglycosylation
might enhance the yield of trehalose production from starch
(7). We also found that decreasing hydrophobic interactions
between enzyme and substrate at positions near subsite +1 led
to the decreased ratios of hydrolysis to transglycosylation for
F405Y MTSase, whereas increasing hydrophobic interactions
and/or removing hydrogen bonding resulted in the increased
ratios of hydrolysis to transglycosylation for Y290F, Y367F,
and Y409F MTSases (8). We suggested that F405Y MTSase
might result in a higher yield of trehalose production from starch
when it replaces wild-type MTSase (8). In this study, site-
directed mutagenesis was used to construct six mutations at
positions near subsites +1 and -1. F405M, F405S, and F405W
MTSases were designed to further decrease or increase the
hydrophobic interactions between residue 405 and substrate as
compared to F405Y MTSase. Additionally, F206W, F206Y, and
F207Y MTSases were also constructed to study the hydrophobic
interactions between enzyme and substrate at positions near
subsite -1. In addition to investigating the kinetic parameters
of the wild-type and mutant enzymes on both transglycosylation
and hydrolysis reactions, we also carried out the production of
trehalose from digestion of 10% (w/v) soluble starch at 75 °C
by using wild-type and mutant MTSases, respectively, along
with isoamylase and MTHase.
kcat, KM, and kcat/KM were calculated by fitting the initial rates as a
function of substrate concentration to the Michaelis-Menten equation
using Enzfitter software (Elsevier-Biosoft). The standard errors of these
parameters were obtained from the fitting results. The change of
transition-state binding energy [∆(∆G)] for substrate hydrolysis caused
by the mutation was used to estimate the binding strength of the
substrate in the transition-state complex and was calculated by the
equation ∆(∆G) ) -RTln[(kcat/KM)mut/(kcat/KM)wt], where the subscripts
mut and wt denote mutant and wild-type enzymes, respectively (10).
Production of Trehalose from Starch. To produce trehalose from
starch, purified thermophilic isoamylase, MTSase, and MTHase were
used concurrently. The purified thermophilic isoamylase and MTHase
were prepared according to the method described in our previous studies
(11, 12). The reaction was carried out at 75 °C with 10% (w/v) soluble
starch in 50 mM citrate phosphate buffer at pH 5. The enzyme
concentration was 0.041 µM for both wild-type and mutant MTSases.
Aliquots of 800 U/g starch of isoamylase and 80 U/g starch of MTHase
were also added into the reaction mixture. Samples were taken at various
time intervals from 0 to 36 h to detect the produced G1 and trehalose.
The reactions were stopped by incubating the mixture in a boiling water
bath for 10 min. For the detection of G1, 0.4 volume of 4 M Tris-HCl
buffer was added to the cooled mixture and the amount of G1 was
measured by a G1 oxidase method (13). For the detection of trehalose,
glucoamylase was added to the cooled mixture at a final concentration
of 5 U/mL to hydrolyze the residual starch, maltooligosaccharides, and
maltooligosyltrehaloses, and the reaction was carried out at 40 °C
overnight. The reaction mixtures were then filtered through a Microcon
Centrifugal Filter Unit with YM-3 membrane (MWCO 3000) to remove
the enzyme. The filtrates were analyzed by a Vercopak Nucleosil 5
µm NH2 column (4.6 mm × 250 mm) with a mobile phase consisting
of 30% (v/v) H2O and 70% (v/v) acetonitrile. Initial rates of trehalose
and G1 formation, respectively, were determined by fitting experimental
data obtained within 1 h to the equation c ) At + B, where c is the
product concentration, t is the time, and A (the initial rate) and B are
obtained by linear regression. The peak trehalose yield was the average
of the three highest consecutive trehalose yield measurements.
MATERIALS AND METHODS
Materials. Escherichia coli Rosetta (DE3) was from Novagen
(Madison, WI). A QuickChange XL Site-Directed Mutagenesis Kit was
purchased from Stratagene (La Jolla, CA). G1, maltose (G2), maltotriose
(G3), maltotetraose (G4), maltopentaose (G5), maltohexaose (G6),
maltoheptaose (G7), glucoamylase, 3,5-dinitrosalicyclic acid (DNS), and
bovine serum albumin (BSA) were supplied by Sigma (St. Louis, MO).
Q-Sepharose, Sephacryl S-200 HR, and protein low molecular weight
standards were from Amersham Pharmacia Biotech (Piscataway, NJ).
Microcon Centrifugal Filter Unit was obtained from Millipore (Bedford,
MA).
Site-Directed Mutagenesis. The MTSase gene in the previously
constructed vector pET-15b-∆H-treY (7) was mutated by polymerase
chain reaction according to the instructions of QuickChange XL Site-
Directed Mutagenesis Kit as previously described (8). The designed
mutations were included in the primers as listed in Table 1.
Production and Purification of MTSase. Wild-type and mutant
MTSases were produced by culturing pET-15b-∆H-treY-transformed
RESULTS
Purity and Thermostability of Wild-Type and Mutant
MTSases. The purity of the enzyme was analyzed by sodium
dodecyl sulfate-polyacrylamide gel (SDS-PAGE; 10% minigel)
according to the method of Laemmli (14). The purity of wild-

5590 J. Agric. Food Chem., Vol. 55, No. 14, 2007
Fang et al.
Table 2. Kinetic Parameters of Wild-Type and Mutant MTSases for Hydrolysis of G3 and Transglycosylation of G4−G7 at 60 °C in 50 mM Citrate
Phosphate Buffer, pH 5
hydrolysis
G3
transglycosylation
MTSase form
wild-typea
G4
G5
G6
G7
-
1
kcat (s
)
44.0
41.3
1.06
±
±
±
1.3b
3.0
0.05
138
±
±
±
4
0.9
0.4
364
±
±
±
7
359
±
±
±
13
0.41
4.7
383
±
±
±
14
0.38
6.6
K
k
M (mM)
13.2
10.4
10.2
5.94
61.2
1.73
0.27
0.8
4.53
79.2
1.34
3.88
98.8
1.07
-
-1
cat/KM (s 1 mM
)
selectivity ratioc
F206W
(×
102)
-
1
kcat (s
)
17.0
14.5
1.18
±
±
±
0.3
0.8
0.05
11.1
1.93
5.78
20.4
7.24
±
±
±
0.1
0.06
1.03
K
k
M (mM)
-
-1
cat/KM (s 1 mM
)
selectivity ratio (
∆
F206Y
×
102)
(
∆
G)d (kJ/mol)
−
0.13
-
1
kcat (s
)
76.3
48.5
1.57
±
±
±
1.7
2.5
0.05
6.11
2.48
2.47
63.6
9.60
±
±
±
0.22
0.23
0.15
K
k
M (mM)
-
-1
cat/KM (s 1 mM
)
selectivity ratio (
G) (kJ/mol)
×
102)
∆
(
∆
−
0.92
F207Y
-
1
kcat (s
)
3.08
61.4
0.050
±
±
±
0.11
4.7
0.002
14.0
2.92
4.80
1.04
7.76
±
±
±
0.3
0.16
0.17
K
k
M (mM)
-
-1
cat/KM (s 1 mM
)
selectivity ratio (
G) (kJ/mol)
×
102)
∆
(
∆
8.63
F405M
-
1
kcat (s
)
44.3
56.1
0.79
±
±
±
2.0
5.6
0.04
196
19.0
10.4
7.60
0.00
±
±
±
11
2.0
0.6
481
6.60
72.9
1.08
±
±
±
9
414
4.63
89.4
0.88
±
±
±
14
0.46
6.0
381
3.38
113
±
±
±
21
0.42
9
K
k
M (mM)
0.31
2.2
-
-1
cat/KM (s 1 mM
)
selectivity ratio (
G) (kJ/mol)
×
102)
0.70
0.37
∆
(
∆
0.81
−
0.48
−0.34
−
F405S
-
1
kcat (s
)
12.4
73.9
0.17
±
±
±
0.4
4.9
0.01
61.0
20.2
3.02
5.63
3.42
±
±
±
2.6
1.6
0.13
197
9.25
21.3
0.80
2.92
±
±
±
10
169
3.75
45.2
0.38
1.55
±
±
±
6
0.40
3.3
147
2.84
51.9
0.33
1.78
±
±
±
4
0.22
2.9
K
k
M (mM)
0.86
1.0
-
-1
cat/KM (s 1 mM
)
selectivity ratio (
G) (kJ/mol)
F405W
×
102)
∆
(
∆
5.07
-
1
kcat (s
)
72.4
53.5
1.35
±
±
±
2.5
4.4
0.07
139
20.6
6.73
20.1
1.20
±
±
±
7
2.0
0.33
410
7.90
51.9
2.60
0.46
±
±
±
19
445
5.90
75.6
1.79
0.13
±
±
±
11
0.37
3.0
429
5.71
83.0
1.63
0.48
±
±
±
6
0.16
1.6
K
k
M (mM)
0.79
2.9
-
-1
cat/KM (s 1 mM
)
selectivity ratio (
G) (kJ/mol)
F405Ya
×
102)
∆
(
∆
−0.67
-
1
kcat (s
)
47.4
54.1
0.876
±
±
±
1.9
4.9
0.050
139
11.6
12.0
7.30
±
±
±
13
2.0
1.1
354
6.34
55.8
1.57
0.26
±
±
±
9
362
4.90
74.0
1.18
0.19
±
±
±
9
0.31
3.1
363
3.80
95.6
0.92
0.09
±
±
±
13
0.37
6.3
K
k
M (mM)
0.45
2.6
-
-1
cat/KM (s 1 mM
)
selectivity ratio (
G) (kJ/mol)
×
102)
∆
(
∆
0.53
−0.40
a Data from ref 8. b Standard error. c Selectivity ratio: [kcat/KM(G3)]/[kcat/KM(Gn)]. d Change of transition-state energy:
∆(∆G) ) −RTln[(kcat/KM)mut/(kcat/KM)wt] (10).
type and mutant MTSases is greater than 95% as indicated by
SDS-PAGE (data not shown). After a 2 h incubation at 80 °C,
the wild-type and mutant MTSases retained virtually all of their
activities (data not shown). The good thermostability of these
enzymes suggested that the purified MTHases were well-folded,
without long-range conformational changes that may be a
consequence of mutation.
decreased rapidly with increasing DP from G4 to G5 and then
dropped slowly beyond that point.
The catalytic efficiencies for hydrolysis and transglycosylation
reactions of F206W MTSase were 110 and 7.3%, respectively,
of those of wild-type MTSase, while those of F206Y MTSase
were 150 and 3.1%, respectively, of those of wild-type MTSase.
However, the catalytic efficiencies for hydrolysis and trans-
glycosylation reactions of F207Y MTSase were only 4.7 and
6.1%, respectively, of those of wild-type MTSase.
Enzyme Kinetics. Kinetic parameters (kcat and KM) for the
hydrolysis of G3 and transglycosylation of G4-G7 at 60 °C and
pH 5 are given in Table 2. F206W, F206Y, and F207Y
MTSases were not very active on transglycosylation as com-
pared with wild-type MTSase; therefore, only G6 was used to
measure the reaction rates for transglycosylation. Except F405S
MTSase, all other MTSases exhibited a previously observed
behavior (8), which kcat values for transglycosylation increased
rapidly with increasing DP from G4 to G5 then to a relatively
constant value with longer substrates. Conversely, KM values
By comparison with wild-type MTSase, F405M MTSase had
slightly higher catalytic efficiencies for the transglycosylation
of G5-G7 and a somewhat lower catalytic efficiency for the
hydrolysis of G3. The catalytic efficiencies of F405S MTSase
for transglycosylation were 29-57% of those of wild-type
MTSase, whereas the catalytic efficiency for hydrolysis was
only 16% of those of wild-type MTSase. F405W MTSase had

Altered Selectivity of Maltooligosyltrehalose Synthase
J. Agric. Food Chem., Vol. 55, No. 14, 2007 5591
Figure 3. Relationship between peak trehalose yields and initial rate of
G1 formation/initial rate of trehalose formation during the digestion of 10%
(w/v) soluble starch at 75 C in 50 mM citrate phosphate buffer (pH 5)
°
using wild-type and mutant MTSases along with isoamylase and MTHase.
formation for F405M and F405Y MTSases as compared with
that of wild-type MTSase. The relationship between peak
trehalose yield (the average of the three highest consecutive
trehalose yield measurements from Figure 2) and the ratios of
initial rates of G1 formation to those of trehalose formation is
plotted in Figure 3. Peak trehalose yields and ratios of initial
rates of G1 formation to those of trehalose formation were
inversely correlated.
MTSase Selectivity toward Hydrolysis vs Transglycosyl-
ation. Selectivity ratios were calculated from the kinetic
parameters of wild-type and mutant MTSases for hydrolysis of
G3 and transglycosylation of G4-G7 as given in Table 2. The
selectivity ratio is expressed as the percentage of catalytic
efficiency of G3 hydrolysis over that of G4-G7 transglycosyl-
ation. F206W and F206Y had significantly increased selectivity
ratios as compared to that of wild-type MTSase, while F405M,
F405S, and F405Y MTSases had decreased selectivity ratios.
The relationships between peak trehalose yield from starch
digestion and the selectivity ratios from kinetic studies for wild-
type and F405M, F405S, F405W, and F405Y MTSases are
plotted in Figure 4. When F405S MTSase was not included,
peak trehalose yields and selectivity ratios were inversely
correlated.
Figure 2. Formation of trehalose during the digestion of 10% (w/v) soluble
starch at 75
and mutant MTSases along with isoamylase and MTHase. Wild-type (
F405M ( ), F405S ( ), F405W ( ), and F405Y ( ).
°
C in 50 mM citrate phosphate buffer (pH 5) using wild-type
b),
O
1
4
9
Table 3. Initial Rates of Trehalose Formation and G1 Formation and
the Ratios of These Two Initial Rates in the Digestion of 10% (w/v)
Soluble Starch Using Wild-Type and Mutant MTSases along with
Isoamylase and MTHase at 75 °C
initial rate (
µ
M/min)
trehalose
MTSase
form
initial rate G1/
initial rate trehalose
G1
wild-type
F405M
F405S
F405W
F405Y
114
96
114
120
106
±
13a
3
10
10
5
556
479
474
563
541
±
±
±
±
±
25
32
24
34
27
0.21
0.20
0.24
0.21
0.19
±
±
±
±
a Standard error.
catalytic efficiencies in general slightly lower than those of wild-
type MTSase except for G3 hydrolysis.
Production of Trehalose from Starch. The digestion of 10%
(w/v) soluble starch was used to study the yield of trehalose
production by wild-type and mutant MTSases along with
isoamylase and MTHase at both a high reaction temperature
(75 °C) and a high substrate concentration. All of the reactions
reached more than 80% trehalose yield in 10 h (Figure 2). The
productivities of trehalose for wild-type and F405M, F405S,
F405W, and F405Y MTSases calculated from the initial rates
of trehalose production were 11.8 ( 0.5, 10.2 ( 0.7, 10.1 (
0.5, 12.0 ( 0.7, and 11.5 ( 0.6 g/L/h, respectively. Because
both MTSase and MTHase catalyze a side reaction to hydrolyze
maltooligosaccharides to release G1, the G1 formation in the
digestion of starch was also determined. The initial rates of
trehalose and G1 formation in the digestion of soluble starch
and their ratios are given in Table 3. Ratios of the initial rates
of G1 formation to trehalose formation were calculated to
evaluate the selectivity for G1 formation vs trehalose formation.
F405W and F405Y MTSases had similar initial rates for
trehalose formation, while F405M and F405S MTSases had
slightly decreased initial rates. F405 M and F405Y MTSases
had slightly lower initial rates for G1 formation as compared
with that of wild-type MTSase, while F405S and F405W
MTSases had similar initial rates. This led to the slightly lower
ratios of initial rates of G1 formation to those of trehalose
Mutations on residue 405 were designed to further alter the
hydrophobic interactions between residue 405 and substrate. The
hydrophobic scale for the side chains of tryptophan, phenyla-
lanine, methionine, tyrosine, and serine were 2.9, 2.3, 2.3, 1.6,
and 0.2 kcal/mol, respectively, according to Karplus (15), and
those were 3.06, 2.43, 1.67, 1.31, and -0.05 kcal/mol, respec-
tively, according to Fauchere and Pliska (16). The relationships
between hydrophobicity of residue 405 and the selectivity ratios
from kinetic studies for wild-type and F405M, F405S, F405W,
and F405Y MTSases are plotted in Figure 5. Except for the
hydrophobicity of methionine obtained from Karplus (15),
hydrophobicities of residue 405 and selectivity ratios were
linearly correlated. In addition, peak trehalose yields and
hydrophobicities of residue 405 were inversely correlated except
for F405S MTSase (Figure 6).
DISCUSSION
In addition to enzyme kinetics, we investigated the effects
of MTSase mutations on trehalose yield by using 10% (w/v)
soluble starch as substrate and combined isoamylase and
MTHase in the reaction mixture. We have previously suggested
that mutant F405Y MTSase with a decreased ratio of hydrolysis

5592 J. Agric. Food Chem., Vol. 55, No. 14, 2007
Fang et al.
Figure 4. Relationships between peak trehalose yields from starch
digestion and the selectivity ratios from kinetic studies by wild-type and
mutant MTSases: (a) selectivity ratios for G3 hydrolysis over G4
transglycosylation, (b) selectivity ratios for G3 hydrolysis over G5
transglycosylation, (c) selectivity ratios for G3 hydrolysis over G6
transglycosylation, and (d) selectivity ratios for G3 hydrolysis over G7
transglycosylation.
Figure 5. Relationships between hydrophobicities of residue 405 and the
selectivity ratios from kinetic studies for wild-type and mutant MTSases:
(a and e) selectivity ratios for G3 hydrolysis over G4 transglycosylation,
(b and f) selectivity ratios for G3 hydrolysis over G5 transglycosylation, (c
and g) selectivity ratios for G3 hydrolysis over G6 transglycosylation, and
(d and h) selectivity ratios for G3 hydrolysis over G7 transglycosylation,
where the hydrophobicity values of the left and the right panels were
based on the estimations of Karplus (15) and Fauchere and Pliska (16),
respectively. F, wild-type; M, F405M; S, F405S; W, F405W; and Y, F405Y.
to transglycosylation might result in a higher yield of trehalose
production from starch when it replaces wild-type MTSase. This
hypothesis, however, has not been tested directly. In this study,
we found that mutant MTSases with decreased selectivity ratios
for hydrolysis over transglycosylation indeed had a higher peak
trehalose yield except for F405S MTSase. However, the ratios
of the initial rate of G1 formation to that of trehalose formation
from the digestion of soluble starch provide a better prediction
for high trehalose yield.
of transition-state energy, ∆(∆G), associated with the loss of a
hydrogen bond between uncharged groups on the substrate and
enzyme, is between 2.1 and 6.3 kJ/mol, while that for loss of a
hydrogen bond between an uncharged group on the substrate
and a charged group on the enzyme is between 14.6 and 18.8
kJ/mol (17, 18).
In this study, the large changes in ∆(∆G) for G3 hydrolysis
and for G6 transglycosylation by F207Y MTSase, 8.63 and 7.76
kJ/mol, respectively, and for G6 transglycosylation by F206W
and F206Y MTSases, 7.24 and 9.60 kJ/mol, respectively (Table
2), are larger than the loss of an uncharged hydrogen bond
between the enzyme and the substrate in the transition state.
However, ∆(∆G) values for G3 hydrolysis by F206W and
F206Y MTSase were only -0.13 and -0.92 kJ/mol, respec-
tively. Because residues F206 and F207 have no abilities of
forming hydrogen bonds, the mutations presumably caused
losses of uncharged hydrogen bonds between the enzyme and
the specified substrate in the neighborhood of mutated residues.
In addition, the losses of hydrogen bonds caused by mutations
on F206 and F207 were different. The losses of hydrogen bonds
caused by mutation on F207 occurred between subsites +1 to
-2 and were observed for the large changes in ∆(∆G)for G3
hydrolysis and for G6 transglycosylation, whereas the losses of
hydrogen bonds caused by mutations on F206 occurred between
subsites -3 to -5 and were observed for the large changes in
∆(∆G) only for G6 transglycosylation, not for G3 hydrolysis.
Residues 206 and 207 were mutated to study how the
hydrophobic interactions between enzyme and substrate at
positions near subsite -1 affect MTSase selectivity. The
mutations on the above residues including F206W, F206Y, and
F207Y are all very conservative mutations. Mutation F206W
was designed to have an increased hydrophobic interaction
between MTSase and substrate, while F206Y and F207Y were
designed to have a decreased hydrophobic interaction. Decreased
hydrophobic interactions between residue F207 and substrate
had decreased selectivity ratio of hydrolysis over transglyco-
sylation. However, either decreased or increased hydrophobic
interactions between residue F206 and substrate had a signifi-
cantly increased selectivity ratio of hydrolysis over transgly-
cosylation.
In an effort to examine whether the alterations in hydropho-
bicity would change the bindings between enzyme and substrate,
we also calculated the transition-state energy to estimate the
binding strength of the enzyme-substrate complex in the
transition state. As some earlier studies pointed out, the change

Altered Selectivity of Maltooligosyltrehalose Synthase
J. Agric. Food Chem., Vol. 55, No. 14, 2007 5593
MTSase was combined with isoamylase and MTHase. Altering
hydrophobic interactions near subsite -1 of the enzyme-
substrate complex could also change MTSase selectivity;
however, either decreased or increased hydrophobic interactions
led to significantly increased selectivity ratios. In addition, the
ratios of the initial rate of G1 formation to that of trehalose
formation from the digestion of soluble starch provide a better
prediction for high trehalose yield than do the selectivity ratios.
ABBREVIATIONS USED
MTSase, maltooligosyltrehalose synthase; MTHase, malto-
oligosyltrehalose trehalohydrolase; G1, glucose; G2, maltose; G3,
maltotriose; G4, maltotetraose; G5, maltopentaose; G6, malto-
hexaose; G7, maltoheptaose; DNS, 3,5-dinitrosalicylic acid;
BSA, bovine serum albumin; ∆(∆G), change of transition-state
energy; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide
gel electrophoresis.
LITERATURE CITED
(1) Richards, A. B.; Krakowka, S.; Dexter, L. B.; Schmid, H.;
Wolterbeek, A. P.; Waalkens-Berendsen, D. H.; Shigoyuki, A.;
Kurimoto, M. Trehalose: A review of properties, history of use
and human tolerance, and results of multiple safety studies. Food
Chem. Toxicol. 2002, 40, 871-898.
(2) Kato, M. Trehalose production with a new enzymatic system
from Sulfolobus solfataricus KM1. J. Mol. Catal. B: Enzym.
1999, 6, 223-233.
(3) Kobayashi, K.; Komeda, T.; Miura, Y.; Kettoku, M.; Kato, M.
Production of trehalose from starch by novel trehalose-producing
enzymes from Sulfolobus solfataricus KM1. J. Ferment. Bioeng.
1997, 83, 296-298.
(4) Mukai, K.; Tabuchi, A.; Nakada, T.; Shibuya, T.; Chaen, H.;
Fukuda, S.; Kurimoto, M.; Tsujisaka, Y. Production of trehalose
from starch by thermostable enzymes from Sulfolobus acidocal-
darius. Starch-Staerke 1997, 49, 26-30.
(5) Kato, M.; Takehara, K.; Kettoku, M.; Kobayashi, K.; Shimizu,
T. Subsite structure and catalytic mechanism of a new glyco-
syltrehalose-producing enzyme isolated from the hyperthermo-
philic archaeum, Sulfolobus solfataricus KM1. Biosci., Biotech-
nol., Biochem. 2000, 64, 319-326.
(6) Kobayashi, M.; Kubota, M.; Matsuura, Y. Refined structure and
functional implications of trehalose synthase from Sulfolobus
acidocaldarius. J. Appl. Glycosci. 2003, 50, 1-8.
(7) Fang, T. Y.; Hung, X. G.; Shih, T. Y.; Tseng, W. C. Charac-
terization of the trehalosyl dextrin-forming enzyme from the
thermophilic archaeon Sulfolobus solfataricus ATCC 35092.
Extremophiles 2004, 8, 335-343.
(8) Fang, T. Y.; Tseng, W. C.; Chung, Y. T.; Pan, C. H. Mutations
on aromatic residues of the active site to alter selectivity of the
Sulfolobus solfataricus maltooligosyltrehalose synthase. J. Agric.
Food Chem. 2006, 54, 3585-3590.
(9) Bradford, M. M. A rapid and sensitive method for the quantitation
of microgram quantities of protein utilizing the principle of
protein-dye binding. Anal. Biochem. 1976, 72, 248-254.
(10) Wilkinson, A. J.; Fersht, A. R.; Blow, D. M.; Winter, G. Site-
directed mutagenesis as a probe of enzyme structure and
catalysis: Tyrosyl-tRNA synthetase cysteine-35 to glycine-35
mutation. Biochemistry 1983, 22, 3581-3586.
(11) Fang, T. Y.; Tseng, W. C.; Yu, C. J.; Shih, T. Y. Characterization
of the thermophilic isoamylase from the thermophilic archaeon
Sulfolobus solfataricus ATCC 35092. J. Mol. Catal. B: Enzym.
2005, 33, 99-107.
(12) Fang, T. Y.; Tseng, W. C.; Guo, M. S.; Shih, T. Y.; Hung, X.
G. Expression, purification, and characterization of the malto-
oligosyltrehalose trehalohydrolase from the thermophilic ar-
chaeon Sulfolobus solfataricus ATCC 35092. J. Agric. Food
Chem. 2006, 54, 7105-7112.
(13) Rabbo, E.; Terkildsen, T. C. Enzymic determination of blood
glucose. Scand. J. Clin. Lab. InVest. 1960, 12, 402-407.
Figure 6. Relationships between hydrophobicities of residue 405 and peak
trehalose yields from starch digestion for wild-type and mutant MTSases:
(a) hydrophobicity values based on the estimations of Karplus (15) and
(b) hydrophobicity values based on the estimations of Fauchere and Pliska
(16).
F405S MTSase also showed a large change in ∆(∆G) for G3
hydrolysis, while the value for transglycosylation decreased with
increasing DP from G4 to G6. These values are close to the loss
of an uncharged hydrogen bond between the enzyme and the
substrate in the transition state. The losses of hydrogen bonds
caused by mutation F405S occurred between subsites +1 to
-2 and might only occur when G3-G5 are used as substrates.
On the other hand, mutations F405M and F405W showed small
∆(∆G) values, from -0.67 to 1.20 kJ/mol, indicating that these
mutations had only minor effects on transition-state substrate
binding.
Altering the hydrophobicity of an active site could affect the
catalytic activity and/or substrate binding of an enzyme (19-
23). Several mutations near subsite +1 of MTSase, including
Y290F, Y367F, F405Y, and Y409F, had been constructed in
our previous study (8). Our previous results showed that
decreasing hydrophobic interactions between the substrate and
the enzyme led to the decreased ratios of hydrolysis to
transglycosylation for F405Y MTSase, whereas increasing
hydrophobic interactions and/or removing hydrogen bonding
resulted in the increased ratios of hydrolysis to transglycosyl-
ation for Y290F, Y367F, and Y409F MTSases (8). In this study,
residue F405 was mutated to methionine, serine, and tryptophan,
which have different degrees of hydrophobicities, in an effort
to further evaluate the relationship between hydrophobic interac-
tions and selectivity. Indeed, the hydrophobicities of residue
405 and selectivity ratios were linearly correlated in general
(Figure 5), and the peak trehalose yields and hydrophobicities
of residue 405 were inversely correlated except for F405S
MTSase (Figure 6). Because F405Y MTSase has a higher peak
trehalose yield and similar trehalose productivity as compared
to those of wild-type, this mutant might be able to replace wild-
type enzyme in the production of trehalose from starch.
In conclusion, the decreased hydrophobicities near subsite
+1 of MTSase could lead to decreased selectivity ratios for
hydrolysis over transglycosylation and may result in increased
trehalose yield from the digestion of soluble starch when

5594 J. Agric. Food Chem., Vol. 55, No. 14, 2007
Fang et al.
(14) Laemmli, U. K. Cleavage of structural proteins during the
assembly of the head of bacteriophage T4. Nature 1970, 227,
680-685.
(15) Karplus, P. A. Hydrophobicity regained. Protein Sci. 1997, 6,
1302-1307.
(21) Suzukawa, K.; Yamagami, T.; Ohnuma, T.; Hirakawa, H.;
Kuhara, S.; Aso, Y.; Ishiguro, M. Mutational analysis of amino
acid residues involved in catalytic activity of a family 18
chitinase from tulip bulbs. Biosci., Biotechnol., Biochem. 2003,
67, 341-346.
(16) Fauchere, J. L.; Pliska, V. Hydrophobic parameters of amino
acid side chains from the partitioning of N-acetyl-amino-acid
amides. Eur. J. Med. Chem. 1983, 18, 369-375.
(17) Fersht, A. R.; Shi, J. P.; Knill-Jones, J.; Lowe, D. M.; Wilkinson,
A. J.; Blow, D. M.; Brick, P.; Carter, P.; Waye, M. M.; Winter,
G. Hydrogen bonding and biological specificity analysed by
protein engineering. Nature 1985, 314, 235-238.
(22) Li, C. H.; Tu, S. C. Active site hydrophobicity is critical to the
bioluminescence activity of Vibrio harVeyi luciferase. Biochem-
istry 2005, 44, 12970-12977.
(23) Siddiqi, F.; Bourque, J. R.; Jiang, H.; Gardner, M.; St. Maurice,
M.; Blouin, C.; Bearne, S. L. Perturbing the hydrophobic pocket
of mandelate racemase to probe phenyl motion during catalysis.
Biochemistry 2005, 44, 9013-9021.
(18) Fang, T. Y.; Coutinho, P. M.; Reilly, P. J.; Ford, C. Mutations
to alter Aspergillus awamori glucoamylase selectivity. I.
Tyr48Phe49fTrp, Tyr116fTrp, Tyr175fPhe, Arg241fLys,
Ser411fAla and Ser411fGly. Protein Eng. 1998, 11, 119-
216.
(19) Erlandsen, H.; Bjorgo, E.; Flatmark, T.; Stevens, R. C. Crystal
structure and site-specific mutagenesis of pterin-bound human
phenylalanine hydroxylase. Biochemistry 2000, 39, 2208-2217.
(20) Fierke, C. A.; Calderone, T. L.; Krebs, J. F. Functional
consequences of engineering the hydrophobic pocket of carbonic
anhydrase II. Biochemistry 1991, 30, 11054-11063.
(24) Davies, G. J.; Wilson, K. S.; Henrissat, B. Nomenclature for
sugar-binding subsites in glycosyl hydrolases. Biochem. J. 1997,
321, 557-559.
Received for review January 16, 2007. Revised manuscript received
May 8, 2007. Accepted May 11, 2007. This work was supported by
Center for Marine Bioscience and Biotechnology at National Taiwan
Ocean University and Grant NSC-94-2313-B-019-025 from the National
Science Council at Taiwan.
JF0701279