Catalytic reaction mechanism based on α-secondary deuterium isotope effects in hydrolysis of trehalose by European honeybee trehalase

Article abstract of DOI:10.1271/bbb.90447

Trehalase, an anomer-inverting glycosidase, hydrolyzes only α, α-trehalose in natural substrates to release equimolecular β-glucose and α-glucose. Since the hydrolytic reaction is reversible, α, α-[1, 1′-²H]trehalose is capable of synthesis from [1-2H]glucose through the reverse reaction of trehalase. α-Secondary deuterium kinetic isotope effects (α-SDKIEs) for the hydrolysis of synthesized α, α-[1, 1′-²H]trehalose by honeybee trehalase were measured to examine the catalytic reaction mechanism. Relatively high k_H/k_D value of 1.53 for α-SDKIEs was observed. The data imply that the catalytic reaction of the trehalase occurs by the oxocarbenium ion intermediate mechanism. In addition, the hydrolytic reaction of glycosidase is discussed from the viewpoint of chemical reactivity for the hydrolysis of acetal in organic chemistry. As to the hydrolytic reaction mechanism of glycosidases, oxocarbenium ion intermediate and nucleophilic displacement mechanisms have been widely recognized, but it is pointed out for the first time that the former mechanism is rational and valid and generally the latter mechanism is unlikely to occur in the hydrolytic reaction of glycosidases.

Full text of DOI:10.1271/bbb.90447

Biosci. Biotechnol. Biochem., 73 (11), 2466–2473, 2009
Catalytic Reaction Mechanism Based on ꢀ-Secondary Deuterium Isotope Effects
in Hydrolysis of Trehalose by European Honeybee Trehalase
1;2;
1
1
3
Haruhide MORI,1; Jin-Ha LEE,y
Masayuki OKUYAMA, Mamoru NISHIMOTO, Masao OHGUCHI,
*
*
y
Doman KIM,4 Atsuo KIMURA,1; and Seiya CHIBA
1;
1Research Faculty of Agriculture, Hokkaido University, Sapporo 060-8589, Japan
2Dental Science Research Institute, School of Dentistry, 2nd Stage of Brain Korea 21 for School of Dentistry,
Chonnam National University, Gwanju 500-757, South Korea
3Technical Section, Fuji Nihon Seito Co., Shimizu 424-8737, Japan
4School of Biological Sciences and Technology and Institute of Bioindustrial Technology,
Chonnam National University, Gwangju 500-757, South Korea
Received June 24, 2009; Accepted August 17, 2009; Online Publication, November 7, 2009
[doi:10.1271/bbb.90447]
Trehalase, an anomer-inverting glycosidase, hydro-
ꢀ; ꢀ-trehalose þ H2O ꢀ ꢁ-glucose þ ꢀ-glucose
lyzes only ꢀ,ꢀ-trehalose in natural substrates to release
equimolecular ꢁ-glucose and ꢀ-glucose. Since the hydro-
lytic reaction is reversible, ꢀ,ꢀ-[1,10-2H]trehalose is
capable of synthesis from [1-2H]glucose through the
reverse reaction of trehalase. ꢀ-Secondary deuterium
kinetic isotope effects (ꢀ-SDKIEs) for the hydrolysis of
synthesized ꢀ,ꢀ-[1,10-2H]trehalose by honeybee treha-
lase were measured to examine the catalytic reaction
mechanism. Relatively high kH=kD value of 1.53 for
ꢀ-SDKIEs was observed. The data imply that the
catalytic reaction of the trehalase occurs by the
oxocarbenium ion intermediate mechanism. In addition,
the hydrolytic reaction of glycosidase is discussed from
the viewpoint of chemical reactivity for the hydrolysis
of acetal in organic chemistry. As to the hydrolytic
reaction mechanism of glycosidases, oxocarbenium ion
intermediate and nucleophilic displacement mechanisms
have been widely recognized, but it is pointed out for the
first time that the former mechanism is rational and
valid and generally the latter mechanism is unlikely to
occur in the hydrolytic reaction of glycosidases.
In the hydrolytic reaction of ꢀ,ꢀ-trehalose with
H218O, the 18O atom has been found to be incorporated
into ꢁ-glucose.5) Since the reaction is reversible, it is
possible to synthesize ꢀ,ꢀ-trehalose from ꢁ-glucose and
ꢀ-glucose by the reverse reaction. The substrate speci-
ficity is extremely high and specific to only ꢀ,ꢀ-
trehalose in natural sugars.
While trehalases are widely distributed in micro-
organisms,6,7) plants,8,9) invertebrates,10–12) and verte-
brates,13–15) their physiological functions are poorly
characterized, except that trehalase(s) in flying insects
can be regarded as particularly significant in the rapid
supply of glucose by hydrolyzing trehalose. However,
several trehalases from different sources16) have been
reported to be capable of hydrolyzing specifically
synthetic substrates, for instance, ꢀ-D-glucopyranosyl
ꢀ-D-galactopyranoside, ꢀ-D-xylopyranosyl ꢀ-D-gluco-
pyranoside, and ꢀ-D-glucopyranosyl 6-deoxy-ꢀ-D-glu-
copyranoside. Particularly, Pseudomonas fluorescens
trehalase showed high hydrolytic activity toward ꢀ-D-
glucopyranosyl ꢀ-galactoside more than toward treha-
lose.16) In addition, it has been found that pig kidney
trehalase hydrolyzes ꢀ-glucopyranosyl fluoride to release
ꢁ-glucose,3) and that the double bound of synthetic 3,7-
anhydro-1,2-didioxy-D-gluco-oct-2-enitol is hydrated by
Trichoderma reesei trehalase to give 1,2-didioxy-D-
gluco-octulose,17) although these two synthetic substrates
have no O-glucosidic linkage.
Two types of trehalases, soluble and bound ones, have
been found in the European honeybee,18) Apis mellifera,
and silkworm,19) Bombix mori. In our previous stud-
ies,4,20) the bound-type trehalase was purified from
European honeybees, and the kinetic properties of the
hydrolytic reaction, molecular cloning of the cDNA, and
the gene structure in chromosome were examined.
Various reaction models have been proposed for the
catalytic reaction mechanisms of glycosidases including
Key words: trehalase; ꢀ,ꢀ-[1,10-2H]trehalose; ꢀ-SDKIE;
oxocarbenium ion intermediate; nucleo-
philic double displacement
Trehalase (ꢀ,ꢀ-trehalase, ꢀ,ꢀ-trehalose glucohydro-
lase, EC 3.2.1.28) is classified into two groups on the
basis of primary structures: glycoside hydrolase family
37 (GH37)1) for most trehalases, and family 65 (GH65)
for some acid trehalases together with phosphor-
ylases (Carbohydrate Active Enzyme database, http:
//www.cazy.org).2)
Trehalase is an anomer-inverting (ꢀ ! ꢁ) glucosi-
dase catalyzing the hydrolysis of the ꢀ-glucosidic O-
linkage of ꢀ,ꢀ-trehalose to liberate equimolecular
quantities of ꢁ-glucose and ꢀ-glucose3–5) (Scheme 1):
y
To whom correspondence should be addressed. Atsuo KIMURA, Tel/Fax: +81-11-706-2808; E-mail: kimura@abs.agr.hokudai.ac.jp; Seiya CHIBA,
Tel/Fax: +81-11-706-2808; E-mail: schiba@abs.agr.hokuda.ac.jp
These two authors equally contributed to this study.
*
Abbreviations: 1-HT, [1-1H]D-glucose; 2-HT, [1-2H]D-glucose; GH family, glycoside hydrolase family; ꢀ-SKIEs, ꢀ-secondary kinetic isotope
effects

Catalytic Reaction Mechanism of Trehalase
2467
H
H
HO
OH
H
H
O
H
H
H
H218O
H
OH
HO
H
O
O
OH
O
HO
OH
HO
HO
18OH
OH
H
H
O
H
H
H
H
+
OH
H
H
HO
HO
H
OH
H
HO
H
OH
OH
β-Glucose
α-Glucose
Scheme 1. Hydrolytic Reaction of ꢀ,ꢀ-Trehalose by Trehalase.
developed on a pre-coated plate (Silica gel 60, Merck, Darmstadt,
Germany) with nitromethane, 1-propanol, and water (4:10:3, by
volume), and detection was done by dipping the plate in methanol
containing 0.3% ꢀ-naphtol and 5% sulfuric acid, and then by heating at
120 ꢁC for 5 min.
trehalase, but a reasonable model explaining both
inverting (ꢀ ! ꢁ, ꢁ ! ꢀ) and retaining (ꢀ ! ꢀ,
ꢁ ! ꢁ) enzyme reactions remains to be established.
As for the catalytic mechanisms so far proposed, two
significant models of a nucleophilic double displace-
ment21) and an oxocarbenium ion intermediate22,23)
reaction mechanism are widely accepted,24–29) but it is
difficult to explain the reaction process of inverting
enzymes such trehalase and glucoamylase by the double
displacement mechanism. While a nucleophilic single
displacement mechanism has been proposed with re-
spect to the hydrolytic reactions of these inverting
enzymes, it is unlikely that such a catalytic reaction
occurs as explained below.
Are the catalytic reaction mechanisms of both invert-
ing and retaining glycosidases explainable by a unified
mechanism? Examination of ꢀ-secondary kinetic iso-
tope effects (ꢀ-SKIEs) is considered to provide funda-
mentally important information for understanding of the
catalytic reaction mechanism.28,29)
This paper describes a catalytic reaction mechanism,
the oxocarbenium ion intermediate mechanism, of a
bound-type GH37 trehalase from European honeybees,
based on the data for ꢀ-SKIEs in the hydrolysis of
synthetic ꢀ,ꢀ-[1,10-2H]trehalose. Further, discussion
will be made as to whether oxocarbenium ion inter-
mediate22,23) or nucleophilic double displacement21)
reaction mechanism is convincing as the hydrolytic
reaction of glycosidases.
MS and NMR analyses. The molecular weight of the synthesized
disaccharide was measured by FD-MS, JMS-SX102A (JEOL, Tokyo,
Japan). Analysis for the anomeric proton in the disaccharide, 10 mg/ml
in D2O, was done by NMR using sodium 3-(trimethylsilyl) propionate-
2,2,3,3-d4 as an internal standard, for 1H-NMR, Bruker AMX-500 (500
MHz) and proton-decoupled 13C-NMR, Bruker AMX-500 (125 MHz).
Enzyme assay and kinetic measurement. Measurement of enzyme
activity was done by methods described in our previous papers.4,30)
Initial velocities in the hydrolysis of [1,10-1H]trehalose and [1,10-
2H]trehalose were measured at various concentrations of both
substrates. A reaction mixture (250 ml) containing the enzyme (2 mg)
and substrate in 0.1 M sodium phosphate buffer (pH 6.7) was incubated
at 35 ꢁC for 6 min. The reaction was terminated by heating at 95 ꢁC for
3 min. The liberated [1-1H] and [1-2H]D-glucose in each case was
determined by the Tris-glucose oxidase-peroxidase method30) with a
modification, using Glucose AR-II (Wako Pure Chemical Ind.). Kinetic
parameters, Km and k0, were determined from Lineweaver-Burk plots
where the straight line intersected the axes at ꢀ1=Km and 1=k0, but the
significant figures are represented in Table 1.
Results
Synthesis of ꢀ,ꢀ-[1,10-1H]trehalose and ꢀ,ꢀ-[1,10-
2H]trehalose
Two disaccharides synthesized from [1-1H] and
[1-2H]D-glucose by the honeybee trehalase-catalyzed
reverse reaction were tentatively designated 1-HT and
2-HT, respectively. Preliminary experiments were done
to determine conditions suitable for trehalose synthesis,
using 15, 20, 30, and 35% D-glucose at 30, 35, 40, and
45 ꢁC for 100 h in a reaction mixture of 250 ml,
containing 2 mg of trehalase at pH 6.7. The preferable
reaction conditions were decided to be 35% D-glucose at
45 ꢁC and for 100 h. While the trehalase was observed to
be inactivated at over 40 ꢁC in our previous study,4) it
was thought to be stabilized under a high concentration
of D-glucose. Preparation of 1-HT and 2-HT was done
under the conditions described above: a reaction mixture
(250 ml) containing of trehalase (2 mg) and 87.5 mg of
[1-1H]D-glucose or [1-2H]D-glucose was incubated at
45 ꢁC for 100 h. The enzyme reaction was terminated by
heating in boiling water for 5 min, and then the mixture
was centrifuged to remove the resulting precipitate. The
supernatant was subjected to a carbon-Celite column,
and the disaccharide fractions, 1-HT and 2-HT, were
individually collected. A trace amount of impurity in the
disaccharide, detected by TLC, was removed by HPLC
of a Hypersil. The result of only 2-HT is shown in
Fig. 1. By repetition of the procedures described above,
about 5 mg of 2-HT was obtained from 1.0 g of [1-2H]D-
glucose.
Materials and Methods
Honeybees and enzymes. European honeybees (Apis mellifera, L.)
were purchased from a beekeeper in Ishikari, Hokkaido, Japan, and
were kept in hives on the campus of Hokkaido University. They were
frozen with liquid nitrogen and stored at ꢀ80 ꢁC and used as occasion
demanded. Trehalase was purified according to a previous paper.4)
Chemicals. [1-1H]D-Glucose (normal D-glucose) and [1-2H]D-
glucose (deuterium-substituted D-glucose) were purchased from Wako
Pure Chemical Industries (Osaka, Japan) and Sigma Chemical (St.
Louis, MO, USA), respectively. ꢀ,ꢀ-Trehalose (1-ꢀ-D-glucopyranosyl
ꢀ-D-glucopyranoside, [1,10-1H]trehalose) was purchased from Nacalai
Tesque (Kyoto, Japan), and was also synthesized from normal D-
glucose by the trehalase-catalyzed reverse reaction (condensation
reaction).
Isolation of trehalose by chromatographies. [1-1H]D-Glucose or
[1-2H]D-glucose was allowed to react with the trehalase condensation
reaction. Each reaction mixture was separately subjected to carbon-
celite (2:3, by weight) column (1:2 ꢂ 7:0 cm) chromatography, and
then the column was washed with water to remove free D-glucose. Two
fractions of disaccharides synthesized from [1,10-1H] and [1,10-2H]D-
glucose were eluted with 30% ethanol and concentrated. For further
purification, each fraction was subjected to high-performance liquid
chromatography (HPLC) equipped with a Hypersil Hypercarb column
(4:6 ꢂ 100 mm; Hypersil, Runcorn, UK) at a flow rate of 0.5 ml per
min with 3% acetonitrile. The purities of isolated trehaloses were first
checked by thin-layer chromatography (TLC): sugar samples were

2468
H. MORI et al.
H-2
H-1
Top
Middle
Bottom
5.4 5.2
3.9
3.8
3.7
H (ppm)
3.6
3.5
3.4
3.3
δ
Fig. 3. 1H-NMR Analysis of Synthesized 1-HT and 2-HT.
Top spectrum, commercial trehalose; middle one, 1-HT synthe-
sized from [1-1H]D-glucose; bottom one, 2-HT synthesized from [1-
2H]D-glucose.
G T 1 2 3
Fig. 1. Thin-Layer Chromatogram of Synthesized 2-HT.
G, [1-2H]glucose; T, commercial ꢀ,ꢀ-trehalose; 1, reaction
mixture of [1-2H]D-glucose and trehalase; 2, 2-HT fraction isolated
by carbon-Celite column; 3, 2-HT isolated by HPLC.
C-1
C-2
C-4
C-5
C-3
C-1 (αC1)
C-6
(βC1)
97
96
94
95
93
100
Fig. 2. 13C-NMR Analyses of Synthesized 1-HT and 2-HT.
90
80
70
δC (ppm)
Upper spectrum, 1-HT synthesized from [1-1H]glucose; lower spectrum, 2-HT synthesized from [1-2H]glucose; inset, enlarged spectrum in
92–98 ppm in upper and lower one. The arrow indicates a chemical shift in anomeric carbon.
Structural analyses of 1-HT and 2-HT
Measurement of ꢀ-secondary deuterium kinetic iso-
tope effects (ꢀ-SDKIEs)
The 1-HT and 2-HT are nonreducing sugars, and were
hydrolyzed by the trehalase to release D-glucose. The
molecular weights of 1-HT and 2-HT as measured by
FD-MS were m=z: 365 ½M þ Naꢃþ and m=z: 367
½M þ Naꢃþ, respectively. This indicates that the mo-
lecular weights are 342 for 1-HT and 344 for 2-HT, and
that the former is a disaccharide consisting of [1-1H]D-
glucose, and the latter, one consisting of [1-2H]D-
glucose. On 13C-NMR analysis of 1-HT and 2-HT, the
anomeric configurations of two glucosyl moieties were
found to be ꢀ-forms (Fig. 2), and the C-1 signal of 2-HT
greatly decreased. Six signals were observed in 1-HT
The ꢀ-SDKIEs for the hydrolysis of commercial
ꢀ,ꢀ-trehalose and 2-HT by the trehalase were examined.
Lineweaver-Burk plots for the effects of the substrate
concentrations (s) versus the initial velocities (v) are
shown in Fig. 4, in which the v values for the hydrolysis
of 2-HT were found to be much lower than those of
ꢀ,ꢀ-trehalose. Table 1 summarizes the rate parameters,
the Michaelis constants (Km), the molecular activities
(kH and kD), and the ratio kH=kD in the hydrolysis of the
two substrates. The ratio kH=kD value for honeybee
trehalase was quite high as compared with the values so
far reported for ꢀ-SDKIEs and ꢀ-ST(tritium)KIEs in the
hydrolytic reaction by other glycosidases. In general,
the values are to the extent of less than 1.3. For instance,
the values for ꢀ-SDKIE range from 1.11 to 1.26 in
the hydrolysis of [1,10-2H]isomaltose,31) p-chlorophenyl
ꢀ-[1-2H]glucoside,32) and ꢀ-[1-2H]glucosyl fluoride33,34)
by ꢀ-glucosidases (retaining enzymes), glucoamylases
(inverting ones), and glucodextranase (an inverting one).
The kH=kD values for ꢀ-STKIEs range from 1.17 to 1.26
1
and five ones, in 2-HT. As shown in Fig. 3, H-NMR
analyses of synthesized 1-HT and a commercial treha-
lose displayed the characteristic JH1 and JH2 values
(3.4 Hz) indicating that the glucosyl configuration was
ꢀ-forms, but no anomeric proton signal was found in
2-HT. These results indicate that 1-HT and 2-HT are
ꢀ,ꢀ-[1,10-1H]trehalose and ꢀ,ꢀ-[1,10-2H]trehalose, re-
spectively.

Catalytic Reaction Mechanism of Trehalase
2469
0.006
0.004
(A)
(B)
Zero-point
level of C1-H
Activated
state
Activated
state
EH < ED(T)
EH = ED(T)
Zero-point
level of C1-D(T)
Zero-point
level of C1-D(T)
EH
EH
ED(T)
ED(T)
Zero-point
level of C1-H
Zero-point
level of C1-D(T)
0.002
Ground state
Ground state
Internuclear distance
Fig. 5. Schematic Curves of Potential Energy Levels vs. Internuclear
Distance in the Initial State and the Activated State of Glycoside
Molecules Having C1-1H and C1-2H(D) or C1-3H(T).
-2
-1
0
1
2
3
4
A, the zero-point energy levels for EH and ED(T) in the activated
state are equivalent or extremely close together; ꢀ-secondary kinetic
isotope effects (ꢀ-SKIEs) are observed, because of EH < ED(T). B,
the difference between EH and ED(T) levels in the ground state and
activated state are equivalent; ꢀ-SKIEs are not observed, because of
1/s, (mM)-1
Fig. 4. Lineweaver-Burk Plots for Hydrolysis of ꢀ,ꢀ-Trehalose and
ꢀ,ꢀ-[1,10-2H]Trehalose by Honeybee Trehalase.
Solid circles ( ), the hydrolysis of commercial ꢀ,ꢀ-trehalose;
open circles ( ), the hydrolysis of ꢀ,ꢀ-[1,10-2H]trehalose. The two
curves were drawn on the assumption that they intersect at one point
on the abscissa. Regrettably, a serious error was recently found in
the scale on the vertical line of Lineweaver-Burk plots in Fig. 4 of
our previous paper,4) in which the graduation was erroneously
depicted at 10-fold. The error is revised in this figure, and the k0
value 86.2 sꢀ1 for hydrolysis of ꢀ,ꢀ-trehalose in the previous paper4)
was consequently calculated to be 860 sꢀ1 (Table 1).
EH ¼ ED(T)
.
of the latter. As shown in step (c) of Scheme 5, the
occurrence of a stable oxocarbenium ion intermediate, a
resonance form between oxonium ion and carbocation
essentially generated by cleavage of C1-O bond, causes
a delay in the hydrolysis rate (kD or T). The kH=kD or T
value is bigger than 1 (unity). This kinetic phenomenon
is called ꢀ-secondary kinetic isotope effects.
Table 1. Kinetic Parameters for the Hydrolysis of ꢀ,ꢀ-[1,10-
1H]Trehalose and ꢀ,ꢀ-[1,10-2H]Trehalose by Trehalase
Figure 5 shows the schematic curves of the potential
energy at the initial stage and the activated stage of
substrate molecule. As shown in Fig. 5A, when ꢀ-
SD(T)KIEs are observed, the difference in energy zero-
point energy levels (EH and ED or T) between the initial
state and the transition state (activated state) is EH <
ED or T, that is, kH > kD or T. The concept of secondary
kinetic isotope effects is explained to be fundamentally
the same as that of primary kinetic isotope effects.28,38,39)
On the other hand (Fig. 5B), the energy zero-point
energy level in the initial state and the transition state
between the normal substrate and the substituted one
are equivalent, EH ¼ ED or T, that is, kH ¼ kD or T. Thus,
ꢀ-SKIEs are not observed. Besides the cases of EH <
ED or T and EH ¼ ED or T, a special case of EH > ED or T
is conceivable, that is, inverse kinetic isotope effects.39)
In the case, kH=kD or T is less than unity, but it seems
unnecessary to take this into consideration in the
reaction of glycosidase.
Enzyme
Substrate
Km (mM)
k0 (sꢀ1
)
kH=kD
Trehalase
[1,10-1H]Trehalose
[1,10-2H]Trehalose
0.66
0.66
860
560
1.53
in the hydrolysis of ꢀ-[1-3H]glucosyl fluoride33,34) by
glucoamylase and glucodextranase. Also in ꢁ-glucosi-
dase, the values for ꢀ-SDKIEs range from 1.10 to 1.12
in the hydrolysis of nitrophenyl ꢁ-[1-2H]glucosides,35)
and from 1.11 to 1.19 in the hydrolysis of phenyl- or
p-nitrophenyl ꢁ-[1,10-2H]chitobiose36) by lysozyme. The
value for ꢀ-STKIEs was 1.19 in the hydrolysis of
[1,10,100-3H]chitotriose37) by lysozyme. The reason for
the relatively high ꢀ-SDKIEs value (kH=kD, 1.53) in the
hydrolytic reaction of [1,10-2H]trehalose by the trehalase
remains unknown, but a possibility may be some effect
of the substitution by deuterium at C1-carbon on
aglycone side.
The kH=kD values for the hydrolysis of honeybee
trehalase, including ꢀ- and ꢁ-glycosidase mentioned
above, confirm that their hydrolytic reaction occurs via
an oxocarbenium ion intermediate catalytic mechanism.
Meanwhile, SDKIEs of non-enzymatic reactions in
acid-catalyzed hydrolysis have been examined using
methyl ꢀ- and ꢁ-glucosides substituted by deuterium at
C1-, C2-, and C5-hydrogen atoms of glycones, kinetic
isotope effects of which are designated ꢀ-, ꢁ-, and ꢂ-
SDKIEs, respectively.29,40) In the non-enzymatic reac-
tions of methyl ꢀ- and ꢁ-glucosides (solvolysis, 2.0 M
HClO4 at 80 ꢁC), ꢀ- and ꢁ-SDKIEs for methyl ꢀ-
glucoside were observed to be kH=kD values 1.13 and
1.107; those for methyl ꢁ-glucoside were kH=kD 1.09 and
1.05, respectively, but no ꢂ-SDKIEs were noticed in
either substrate. These results seem to be of importance in
understanding the reaction mechanism of glycosidases.
The schematic relation of free energy levels during
reaction progresses, which is mutually related to the
potential energy levels of the ground and activated states
Discussion
ꢀ-Secondary deuterium (D) and tritium (T) kinetic
isotope effects, ꢀ-SD(T)KIEs, have been observed in the
hydrolytic reaction of many ꢀ- and ꢁ-glycosidases,
including anomer retaining and inverting enzymes. The
concept of ꢀ-SD(T)KIEs is to be interpreted as follows.
In the hydrolytic reaction of the substituted glycoside by
D or T at C1-H and the unsubstituted substrate, there is
the case that the rate (kD or T) of the hydrolysis of the
former becomes lower than that (kH) of the hydrolysis

2470
H. MORI et al.
O
O
O
O
O
O
H
H
H
H
H
H
H
OH
H-O-R
OH
HO
O
H
O
OH
O
O
HO
O
H
HO
H
H
H
H
H
H
O
H
HO
H
OH
HO
HO
OH
OH
H
O
O
H
H
H
R
R
H
O
H
O
O
O
O
O
(a)
(b)
(c)
Scheme 2. Nucleophilic Single Displacement Reaction Mechanism.
The C1, C1–H, C2, C5, and ring-O exist on a plane of sp2 hybrid orbital in a transition state b.
‡
∆G1‡ < ∆G2
T2
‡
(C)
∆G1‡ = ∆G2
(B)
(A)
T1
T1
T1
‡
∆G2
T2
‡
∆G1‡ > ∆G2
‡
∆G2
‡
‡
‡
∆G1
∆G1
∆G1
C⊕
G–E
∆G°
∆G°
∆G°
Reaction progress
Fig. 6. Schematic Change of Free Energy during the Reaction in Hydrolysis of Glycoside by Glycosidase.
A, the process in the case of oxocarbenium ion intermediate mechanism. B, the process in the case of nucleophilic single displacement
mechanism. C, the process in the case of nucleophilic double displacement mechanism. T1 and T2 indicate the first and second transition states,
respectively. Cꢄ indicates the energy level of oxocarbenium ion intermediate, and G-E, that of glycosyl-enzyme intermediate.
(Fig. 5), is shown in Fig. 6. Figures 6B and C indicate
nucleophilic single and double displacement mecha-
nisms, respectively, in which ꢀ-SDKIEs are not ob-
served in the reaction processes as mentioned above, and
Fig. 6A indicates the oxocarbenium ion intermediate
mechanism.
reaction. As Scheme 2 shows, the hydrolytic reaction
is terminated through one time transition state (b),
lacking the process of oxocarbenium ion intermediate.
This means that ꢀ-SKIEs are not observed in the
reaction; nevertheless, ꢀ-SD(T)KIEs have been ob-
served in the hydrolytic reaction of inverting enzymes
such as trehalase and glucoamylase.31,33) At the final
stage c of Scheme 2, the positional states of the catalytic
carboxy group (acid/base) and carboxylate ion (base)
are reversed. The situation of catalytic groups at the
active enzyme must revert to the initial situation of stage
a. If not, the reversible synthesis of trehalose shown in
Scheme 1 is impossible. These findings cannot be
explained by the nucleophilic single displacement
mechanism of Scheme 2.
Similar things can be said as to the nucleophilic double
displacement mechanism in Scheme 3. In the reaction,
ꢀ-SKIEs may not be observed, because the process of
oxocarbenium ion intermediate is lacking. In the process
c via first transition state b, C1-O bond is newly formed
between glycone and carboxyl ion of enzyme to give a
kind of intermediate. The newly formed covalent bond is
split by nucleophilic attack of water on C1-carbon
through process d to f including the second transition
state, e. It is thought that the reaction is a second-order
bimolecular reaction between substrate and enzyme, and
proceeds like SN2 reaction, but the cleavage of C1-O
Nucleophilic single displacement mechanism is ex-
plained as the reaction of nucleophilic attack of water on
C1-carbon and electrophilic attack of Hþ ion of carboxyl
group (acid/base) on oxygen of C1-O bond; simulta-
neously carboxylate ion (base) pulls out Hþ ion of water
(Scheme 2). However, such a hydrolytic reaction is
unlikely to occur continuously, because Hþ ion as the
initiation-driving force essential to the subsequent
hydrolytic reaction is missing from the catalytic group
(acid/base) at the final stage (c). As for the mechanism
of hydrolysis of retaining glycosidases, the nucleophilic
double displacement and the oxocarbenium ion inter-
mediate reaction mechanisms are usually applied, and a
number of investigators are liable to receive the former
mechanism as the reaction mechanism of retaining
enzymes. However, the reaction mechanism of inverting
enzymes is not explicable by the nucleophilic double
displacement mechanism, because ‘‘triple’’ displacement
may become necessary to invert anomer.25) Therefore,
nucleophilic single displacement mechanism41,42) has
been proposed so as to account for the inverting

Catalytic Reaction Mechanism of Trehalase
2471
H
H
OH
H
OH
HO
O
H
OH
HO
O
H
O
O
O
O
O
O
O
H-O-R
HO
H
H2O
H
H
H
H
H
HO
HO
OH
OH
HO
OH
H
H
O
H
O
O
H
R
R
H
H
O
O
O
O
O
(a)
(b)
(c)
H
H
OH
HO
O
H
OH
OH
H
HO
O
H
H
O
O
O
O
O
O
O
O
H
HO
H
HO
H
H
H
OH
HO
HO
OH
H
H
OH
H
O
H
O
H
O
H
H
H
H
H
O
O
O
O
O
(d)
(e)
(f)
Scheme 3. Nucleophilic Double Displacement Reaction Mechanism.
The C1, C1-H, C2, C5, and ring-O also exist on a plane of sp2 hybrid orbital in the two transition states b and e.
bond of the glycosyl-enzyme intermediate probably does
not occur by nucleophilic attack of water on C1-carbon.
As shown in Fig. 6C, ꢀG1 and ꢀG2 at the two
transition states are depicted to be equal as a matter of
convenience. If the energy levels are ꢀG1 > ꢀG2 , the
rate-limiting step may be present in the first transition
state (b in Scheme 3); if the energy levels are
ꢀG1 < ꢀG2 , the rate-limiting step may be present in
the second transition state, e. Therefore, it is practically
difficult to decide the rate-limiting step in the reaction.
The oxocarbenium ion intermediate and the transition
state of double displacement are apparently similar to
the coordination of bonds to carbon in the reactions of
alkyl halides,43,44) for instance, in the nucleophilic
substitution (unimolecular reaction) of 2-bromo-2-meth-
ylpropane and water (Fig. 7A) as a typical SN1 reaction,
and in the nucleophilic substitution (bimolecular reac-
tion) of methyl bromide and hydroxy ion (Fig. 7B) as a
typical SN2 reaction, respectively. The oxocarbenium
ion intermediate and nucleophilic double displacement
reaction mechanisms of the hydrolysis of acetal are
basically different from the nucleophilic substitution
reaction of alkyl halides, and hence are not intrinsic SN1
and SN2 reactions. Of the two hydrolytic reactions for
glycoside, the former should be called SN1-like reaction,
and the latter, SN2-like one.
A great many researchers have accepted the nucleo-
philic single and double displacement mechanisms, but
the mechanisms are considered to comprise inconsis-
tencies that cannot be neglected from the viewpoint of
organic chemistry, as mentioned above. Consequently, it
is inferred that there are actually no reactions dependent
on both nucleophilic single and double displacement
reaction mechanisms.
In order to solve the problem of the reaction
mechanism, another novel approach other than SN1
and SN2 reactions should be attempted. One approach is
to regard the hydrolytic reaction of glycosidic-O bond as
cleavage of acetal (Scheme 4). In the double displace-
ment mechanism (Scheme 3), the two acetals, substrate
Plane of sp2 hybrid orbital
(A)
(B)
Empty 2p orbital
H
CH3
H
H3C
H
H
H
H
-
-
HO
Br
C
O
O
C
-
H
CH3
+ Br
Fig. 7. Carbocation Intermediate and Transition State for the Reac-
tions of Two Alkyl Halides.
A, the carbocation intermediate of nucleophilic substitution of
2-bromo-2-methylpropane and water (SN1 reaction). In this case,
water attacks carbocation from the left or right side. B, the transition
state of nucleophilic substitution of bromomethane and hydroxyl ion
(SN2 reaction). In this case, Walden’s inversion occurs.
a and glycosyl-enzyme c, are cleaved by different action
modes; that is, nucleophilic attack by carboxylate ion
in step (a) and by water on C1-carbon in step (d),
respectively. However, it is generally known that the
initial attack of only Hþ ion on the oxygen of C1-O bond
of acetal is essential to cleavage of the bond.45–50)
Therefore, acetal d in Scheme 3 appears unlikely to be
split by a different system. As shown in Scheme 4, the
most important point for the hydrolytic reaction of acetal
is the fact that oxonium ion is essentially formed by an
attack of Hþ ion on oxygen of C1-O bond.49)
The reasons why many investigators have accepted
the nucleophilic double displacement reaction mecha-
nism are obscure, but a probable reason is attributed to
numerous reports on mechanism-base inhibitors. Avail-
able inhibitors,51,52) conduritol B epoxide,53) 2-deoxy-2-
fluoro-glucoside,54) 1,1-difluoroalkyl-glucosides,55) and
5-fluoro-ꢀ-D-glucopyranosyl fluoride,56) have been syn-
thesized as a novel class of enzyme-activated irrever-

2472
H. MORI et al.
H
O
H+
HOR
H2O
H+
R
O R
O R'
OH2
O R'
OH
C
C
C
O
C
R'
O
R'
C
C
O R'
O
R'
OH2
Acetal
Hemiacetal
Scheme 4. Hydrolytic Reaction of Acetal to Hemiacetal.
In the case of hydrolysis of glycoside, C corresponds to C1 of glycone; R, to aglycone; R’, to C5 forming O-ring.
H
H
OH
H
OH
O
OH
H
HO
O
H
H
H
OH
H
OH
H
O
O
O
O
O
O
HO
O
O
O
O
O
O
O
H-O-R
O
O
HO
H
H
H
H
HO
HO
H
H
H
HO
H
HO
H
H
H
OH
OH
O
HO
OH
HO
HO
H
O
H
OH
OH
R
H
R
H
H
H
O
H
O
O
O
(a)
(b)
(c)
(d)
(e)
H
O
O
H
OH
O
O
H
HO
O
H
H
OH
O
O
H
OH
O
O
O
HO
O
H
H
H
HO
O
H
H
HO
H
H
O
H
OH
H
H
O
HO
H
OH
H
HO
H2O
H
OH
H
H
H
OH
H
O
O
O
O
O
O
HO
H
H
(f)
(g)
(h)
HO
OH
H
H2O
H
H
OH
H
OH
H
HO
O
H
OH
O
H
O
O
O
H
O
O
H
O
H
HO
(e)
HO
H
H
H
H
H
HO
OH
HO
HO
OH
OH
H
O
O
O
H
H
H
H
H
H
O
O
O
O
O
O
(f’)
(g’)
(h’)
Scheme 5. Oxocarbenium Ion Intermediate Reaction Mechanism.
The C1, C1-H, C2, C5, and ring-O exist on a plane sp2 hybrid orbital in transition states b, including c and d, g, and g0. Step e indicates
oxocarbenium ion intermediate. The steps from f to h are the cases of inverting glycosidases like glucoamylase and trehalase, and the steps from
f0 to h0, retaining ones like ꢀ-amylase and ꢀ-glucosidase. The parts of the shadowed boxes indicate the carboxyl groups (base and acid/base)
involved in the catalytic reaction at the active site.
sible inhibitors, which are mechanism-base inhibitors
(suicide substrates). 2-Deoxy-2-fluoro-glucosides, 1,1-
difluoroalkyl-glucosides, and 5-fluoro-ꢀ-glucosyl fluo-
ride are regarded as substrate analogs, and conduritol B
epoxide, as an analog in the transition state.57,58) In
mechanism-base inhibition,59) usually glycone analog
binds to a carboxylate group in active site directly
involved in the activity, and then the glycosidase is
irreversibly inactivated. Mechanism-base inactivation
reactions for many ꢀ- and ꢁ-glucosidases54–58,60–62) and
ꢁ-glucan glucohydrolases63–65) have been examined
using the inhibitors mentioned above. The time-depend-
ence for inactivation showed curves of definite pseudo
first-order reaction. This means that the reaction modes
are dead-end, that is, almost irreversible reactions, and
the glycone analogs have confirmed to be connected to a
carboxylate group in the catalytic site.24,56,66,67) The
identification of such a covalent bond has been regarded
as a proof in support of the nucleophilic double
displacement reaction mechanism. However, even if
the glycone analog is proved to be linked to a carbox-
ylate group essential for catalytic activity, the formation
of glycosyl-enzyme intermediate by covalent-O bond is
not always evidenced in the hydrolytic reaction of
normal substrates. The reactions of substrate analogs
and glycosidases so far examined are just the same as
general mechanism-based modification with suicide
substrates, and accordingly it cannot be considered to
be evidence that the usual hydrolysis of the normal
substrates by glycosidase occurs through double dis-
placement reaction mechanism.
On the other hand, the oxocarbenium ion intermediate
reaction mechanism (Scheme 5) is also generally
known. The schematic course of hydrolysis is shown
in Scheme 5, the processes of which are basically
similar to the cleavage reaction of acetal to hemiacetal
involving the formation of oxonium ion (c). The reaction
mechanism in Scheme 5 is applicable to both inverting
and retaining glycosidases. This means that the reaction
mechanism of glycosidase becomes explicable by a
unitary mechanism. As mentioned above, the double
displacement mechanism is improbable in view of
chemical reactivity. Both the oxocarbenium ion inter-
mediate mechanism and the nucleophilic double dis-
placement mechanism are described in textbooks,26,27)
but the description in a textbook27) has recently been
revised to the effect that the latter mechanism is more
plausible than the former. There has been few reports
discussed from aspects on reactivity in organic chem-
istry. The oxocarbenium ion intermediate reaction
mechanism appears to be a plausible mechanism in
both inverting and retaining glycosidase.

Catalytic Reaction Mechanism of Trehalase
2473
34) Matsui H, Chiba S, and Hehre EJ, Denpun Kagaku, 38, 181–185
(1991).
Acknowledgment
35) Kempton JB and Withers SG, Biochemistry, 31, 9961–9969
(1992).
We are indebted to Dr. A. Murai, Professor Emeritus,
who was Professor of Organic Chemistry in Faculty of
Science of Hokkaido University, for helpful discussion
and advice. We thank also Dr. E. Fukushi and Mr.
K. Watanabe for obtaining the MS, 1H-NMR, and
13C-NMR data.
36) Dahlquist FW, Rand-Meir T, and Raftery MA, Proc. Natl. Acad.
Sci. USA, 61, 1194–1198 (1968).
37) Smith LE, Mohr LH, and Raftery MA, J. Am. Chem. Soc., 95,
7497–7500 (1973).
38) Husky WP, ‘‘Enzyme Mechanism from Isotope Effects,’’ ed.
Cook PF, CRC Press, Inc., Boca Raton, Ann Arbor, Boston,
London, pp. 37–72 (1991).
References
39) Naito S, Kagaku (in Japanese), 50, 169–171 (1995).
40) Bennet AJ and Sinnott ML, J. Am. Chem. Soc., 108, 7287–7294
(1986).
1) Henrissat B and Bairoch A, Biochem. J., 293, 781–788 (1993).
2) Coutinho PM and Henrissat B, ‘‘Recent Advances in Carbohy-
drate Bioengineering,’’ eds. Gilbert HJ, Davies G, Henrissat B,
and Svensson B, The Royal Society of Chemistry, Cambridge,
pp. 3–12 (1999).
41) Wang Q, Graham RW, Trimbur D, Warren RAJ, and Withers
SG, J. Am. Chem. Soc., 116, 11594–11595 (1994).
42) Stam MR, Blanc E, Coutinho PM, and Henrissat B, Carbohydr.
Res., 340, 2728–2734 (2005).
3) Nakano M, Brewer CF, Kasumi T, and Hehre EJ, Carbohydr.
Res., 194, 139–144 (1989).
43) Clayden J, Greeves N, Warren S, and Wothers P, ‘‘Organic
Chemistry,’’ Oxford University Press Inc., New York, pp. 339–
358 (2001).
4) Lee JH, Tsuji M, Nakamura M, Nishimoto M, Okuyama M,
Mori H, Kimura A, Matsui H, and Chiba S, Biosci. Biotechnol.
Biochem., 65, 2657–2665 (2001).
44) McMurry JC, ‘‘Organic Chemistry’’ 6th ed., Brooks/Cole, A
Division of Thomson Learning, Inc., pp. 682–729 (2004).
45) Young PR and Jencks WP, J. Am. Chem. Soc., 99, 8238–8248
(1977).
5) Clifford KH, Eur. J. Biochem., 106, 337–340 (1980).
6) Hey AE and Elbein AD, J. Bacteriol., 96, 105–110 (1968).
7) Vijayakumar P, Ross W, and Reese ET, Can. J. Microbiol., 24,
1280–1283 (1978).
46) Eliason R and Kreevoy MM, J. Am. Chem. Soc., 100, 7037–
7041 (1978).
8) Hutson DH and Manners DJ, Biochem. J., 94, 783–789 (1965).
9) Alexander AG, Plant Cell Physiol., 14, 157–168 (1973).
10) Saito S, J. Biochem., 48, 101–109 (1960).
47) Kresge AJ and Weeks DP, J. Am. Chem. Soc., 106, 7140–7143
(1984).
11) Talbot BG and Huber RE, Insect Biochem., 5, 337–347 (1975).
12) Mitsumasu K, Azuma M, Niimi T, Yamashita O, and Yaginuma
T, Insect Mol. Biol., 14, 501–508 (2005).
48) Amyes TL and Jencks WP, J. Am. Chem. Soc., 111, 7888–7900
(1989).
49) Clayden J, Greeves N, Warren S, and Wothers P, ‘‘Organic
Chemistry,’’ Oxford University Press Inc., New York, pp. 407–
445 (2001).
13) Sacktor B, Proc. Natl. Acad. Sci. USA, 60, 1007–1014 (1968).
14) Nakano M, Sumi Y, and Miyakawa M, J. Biochem., 81, 1041–
1049 (1977).
50) McMurry JC, ‘‘Organic Chemistry’’ 6th ed., Brooks/Cole, A
Division of Thomson Learning, Inc., pp. 343–383 (2004).
´
51) Lalegerie P, Legler G, and Yon JM, Biochimie, 64, 977–1000
(1982).
15) Galand G, Biochim. Biophys. Acta, 789, 10–19 (1984).
16) Labat-Robert J, Baumann FC, Bar-Guilloux E, and Robic D,
Comp. Biochem. Physiol. B, 61, 111–114 (1978).
17) Weiser W, Lehmann J, Chiba S, Matsui H, Brewer CF, and
Hehre EJ, Biochemistry, 27, 2294–2300 (1988).
52) Mooser G, ‘‘The Enzyme’’ 3rd ed., Vol. XX, ed. Sigman DS,
Academic Press, Inc., San Diego, New York, Boston, London,
Sydney, Tokyo, Toronto, pp. 187–233 (1992).
18) Lefebvre YA and Huber RE, Arch. Biochem. Biophys., 140,
514–518 (1970).
53) Braun H, Legler G, Deshusses J, and Semenza G, Biochim.
Biophys. Acta, 483, 135–140 (1977).
19) Gussin AE and Wyatt GR, Arch. Biochem. Biophys., 112, 626–
634 (1965).
54) Withers SG, Street IP, Bird P, and Dolphin DH, J. Am. Chem.
Soc., 109, 7530–7531 (1987).
20) Lee JH, Saito S, Mori H, Nishimoto M, Okuyama M, Kim D,
Wongchawalit J, Kimura A, and Chiba S, Biosci. Biotechnol.
Biochem., 71, 2256–2265 (2007).
55) Halazy S, Danzin C, Ehrhard A, and Gerhart F, J. Am. Chem.
Soc., 111, 3484–3485 (1989).
21) Koshland DE, Biol. Rev. Camb. Philos. Soc., 28, 416–436 (1953).
22) Phillips DC, Proc. Natl. Sci. Acad. USA, 57, 483–495 (1967).
23) Rupley JA and Gates V, Proc. Natl. Sci. Acad. USA, 57, 496–
510 (1967).
56) Lee SS, He S, and Withers SG, Biochem. J., 359, 381–386 (2001).
57) Kimura A, Takata M, Fukushi Y, Mori H, Matsui H, and Chiba
S, Biosci. Biotechnol. Biochem., 61, 1091–1098 (1997).
58) Okuyama M, Okuno A, Shimizu N, Mori H, Kimura A, and
Chiba S, Eur. J. Biochem., 268, 2270–2280 (2001).
59) Nelson DL and Cox MM, ‘‘Lehninger Principles of Biochem-
istry’’ 5th ed., WH Freeman and Company, New York, pp. 183–
233 (2008).
24) Vocadlo DJ, Davies GJ, Laine R, and Withers SG, Nature, 412,
835–838 (2001).
25) Chiba S, Biosci. Biotechnol. Biochem., 61, 1233–1239 (1997).
26) Metzler DE, ‘‘Biochemistry: The Chemical Reactions of Living
Cells’’ 2nd ed., Vol. 1, Harcourt/Academic Press, San Diego,
pp. 589–608 (2001).
60) Withers SG, Rupitz K, and Street IP, J. Biol. Chem., 263, 7929–
7932 (1988).
27) Voet D and Voet J, ‘‘Biochemistry’’ 3rd ed., John Wiley & Sons,
Inc., New York, London, Sydney, Toronto, pp. 496–515 (2004).
28) Kuby SA, ‘‘A Study of Enzymes’’ Vol. 1, CRC Press, Inc., Boca
Raton, Ann Arbor, Boston, London, pp. 262–272 (1990).
29) Schramm VL, ‘‘Enzyme Mechanism for Isotope Effects,’’ ed.
Cook PF, CRC Press, Inc., Boca Raton, Ann Arbor, Boston,
London, pp. 368–381 (1991).
61) Surarit R, Matsui H, Chiba S, Svasti J, and Srisomsap C, Biosci.
Biotechnol. Biochem., 60, 1265–1268 (1996).
62) Fukuda K, Mori H, Okuyama M, Kimura A, Ozaki H,
Yoneyama M, and Chiba S, Biosci. Biotechnol. Biochem., 66,
2060–2067 (2002).
63) Mackenzie LF, Brooke GS, Cutfield JF, Sullivan PA, and
Withers SG, J. Biol. Chem., 272, 3161–3167 (1997).
64) Tull D, Withers SG, Gilkes NR, Kilburn DG, Warren RA, and
Aebersold R, J. Biol. Chem., 266, 15621–15625 (1991).
65) Hrmova M, Varghese JN, De Gori R, Smith BJ, Driguez H, and
Fincher GB, Structure, 9, 1005–1016 (2001).
30) Kimura A, Takewaki S, Matsui H, Kubota M, and Chiba S,
J. Biochem., 107, 762–768 (1990).
31) Igaki S, Kimura A, and Chiba S, Oyo Toshitsu Kagaku (J. Appl.
Glycosci.), 45, 269–274 (1998).
32) Cogoli A and Semenza GA, J. Biol. Chem., 250, 7802–7809
(1975).
66) Withers SG and Street IP, J. Am. Chem. Soc., 110, 8551–8553
(1988).
33) Matsui H, Blanchard JS, Brewer CF, and Hehre EJ, J. Biol.
Chem., 264, 8714–8716 (1989).
67) Withers SG, Warren RAJ, Street IP, Rupitz K, Kempton JB, and
Aebersold R, J. Am. Chem. Soc., 112, 5887–5889 (1990).