Thermostable variants of pyranose 2-oxidase showing altered substrate selectivity for glucose and galactose

Article abstract of DOI:10.1021/jf9040047

The homotetrameric flavoprotein pyranose 2-oxidase (P2Ox) has several proposed biotechnological applications, among others as a biocatalyst for carbohydrate transformations toward higher-value products. To improve some of the catalytic properties of P2Ox from Trametes multicolor, we selected a semirational enzyme engineering approach, namely, saturation mutagenesis of the amino acid His 450 located at a pivotal point of the active site loop and subsequent screening of the libraries thus obtained for improved activity with the sugar substrate D-galactose. A variant with improved catalytic characteristics identified was H450G, which showed a significant, 3.6-fold decrease In K_M together with a 1.4-fold increase in κ_cat for its substrate D-galactose and an overall improvement in the catalytic efficiency by a factor of 5. By combining H450G with other amino acid replacements, we obtained the P2Ox variants H450G/V546C and H450G/E542K/V546C, which can be of interest for applications in food industry due to their increased activity with D-galactose, high activity with D-glucose, and considerably increased stability for the latter variant. While the His-tagged recombinant wild-type enzyme strongly prefers D-glucose to D-galactose as its substrate, H450G/E542K/V546C converts both sugars, which are found in lactose hydrolysates, concomitantly, as was shown by laboratory-scale biotransformation experiments. The 2-keto sugars thus obtained can conveniently be reduced to the corresponding ketoses D-fructose and D-tagatose. 2010 American Chemical Society.

Full text of DOI:10.1021/jf9040047

J. Agric. Food Chem. 2010, 58, 3465–3471 3465
DOI:10.1021/jf9040047
Thermostable Variants of Pyranose 2-Oxidase Showing Altered
Substrate Selectivity for Glucose and Galactose
†,‡
,†
*
O
LIVER
S
PADIUT,† TIEN-THANH
NGUYEN
,
AND
DIETMAR
HALTRICH
†Food Biotechnology Laboratory, Department of Food Science and Technology, BOKU;University of
Natural Resources and Applied Life Sciences, A-1190 Vienna, Austria, and ‡Institute of Biological and
Food Technology, Hanoi University of Technology, Hanoi, Vietnam
The homotetrameric flavoprotein pyranose 2-oxidase (P2Ox) has several proposed biotechnological
applications, among others as a biocatalyst for carbohydrate transformations toward higher-value
products. To improve some of the catalytic properties of P2Ox from Trametes multicolor, we
selected a semirational enzyme engineering approach, namely, saturation mutagenesis of the amino
acid His450 located at a pivotal point of the active site loop and subsequent screening of the
libraries thus obtained for improved activity with the sugar substrate
improved catalytic characteristics identified was H450G, which showed a significant, 3.6-fold
decrease in KM together with a 1.4-fold increase in kcat for its substrate -galactose and an overall
D-galactose. A variant with
D
improvement in the catalytic efficiency by a factor of 5. By combining H450G with other amino acid
replacements, we obtained the P2Ox variants H450G/V546C and H450G/E542K/V546C, which can
be of interest for applications in food industry due to their increased activity with
high activity with -glucose, and considerably increased stability for the latter variant. While the
His-tagged recombinant wild-type enzyme strongly prefers -glucose to -galactose as its substrate,
D-galactose,
D
D
D
H450G/E542K/V546C converts both sugars, which are found in lactose hydrolysates, concomitantly,
as was shown by laboratory-scale biotransformation experiments. The 2-keto sugars thus obtained
can conveniently be reduced to the corresponding ketoses D-fructose and D-tagatose.
KEYWORDS: Pyranose oxidase; hydrolyzed lactose; enzyme engineering; saturation mutagenesis;
D-tagatose; D-fructose; substrate specificity; biocatalytic conversion
INTRODUCTION
become an attractive biocatalyst for biotransformations of car-
bohydrates, as it can be used for the synthesis of various
carbohydrate derivates and rare sugars (9). Among others, the
Pyranose 2-oxidase (P2Ox; pyranose:oxygen 2-oxidoreduc-
tase; glucose 2-oxidase; EC 1.1.3.10) is a flavin adenine dinucleo-
tide (FAD) dependent enzyme widespread in wood-degrading
basidiomycetes (1-3). P2Ox catalyzes the oxidation of various
aldopyranoses at C2 to yield the corresponding 2-keto-sugars
using oxygen as well as several other compounds (quinones, metal
ions) as its electron acceptors. A search in the US Patent data
bank (1976-2009) resulted in 438 patents found under the key
word “pyranose oxidase”, showing the applied interest in this
enzyme. Actual applications of P2Ox can be found in food
technology (e.g., the baking industry), bioelectrochemistry (e.g.,
enzymatic biofuel cells and biosensors (4)) and analytics, where
oxidation of
2-keto- -galactose is of applied interest as these oxidized inter-
mediates can subsequently be reduced at position C-1 to obtain
the ketoses -fructose and -tagatose (10), both of which are of
D-glucose and D-galactose to 2-keto-D-glucose and
D
D
D
interest for the food industry. Lactose, a byproduct in the dairy
industry, can be hydrolyzed conveniently by β-galactosidases
resulting in equal amounts of D-glucose and D-galactose (11),
which can subsequently be used as substrates for P2Ox-catalyzed
conversions. Lactose represents about 4.5-5% (w/v) of whey,
accumulating to approximately 6 million tons worldwide per year
from 145 million tons of liquid whey (12), which has commonly
been considered a waste product. About half of the world’s whey
production is disposed of as effluent, causing considerable
environmental problems due to the high organic content and
large volumes. Direct utilization of lactose in food and pharma-
ceuticals is limited, among others because of the reduced lactase
activity encountered in adults (approximately 70-75% world-
wide). Therefore, alternative ways to exploit the renewable
resource lactose, including enzymatic sugar transformations,
are of considerable commercial interest. For effective biotechno-
the enzyme is used for the determination of 1,5-anhydro-D-
glucitol, a marker for glycemic control in diabetes (5). The first
proposed industrial application of P2Ox was the conversion of
D-glucose to pure D-fructose via the intermediate 2-keto-D-glu-
cose by the company Cetus in the early 1980s (6). In latter studies,
various other mono- and disaccharides were converted to the
corresponding 2-keto-sugars by P2Ox (7,8). Since then, P2Ox has
*To whom correspondence should be addressed. Lebensmittelbio-
€
€
technologie, Universitat fur Bodenkultur, Muthgasse 18, A-1190
Wien, Austria. E-mail: dietmar.haltrich@boku.ac.at. Tel: þ43-1-
47654-6140. Fax: þ43-1-47654-6251.
logical applications, the catalytic activity of P2Ox with
D-galac-
tose, however, is too low (5.7% relative activity compared to
© 2010 American Chemical Society
Published on Web 02/16/2010
pubs.acs.org/JAFC

3466 J. Agric. Food Chem., Vol. 58, No. 6, 2010
-glucose (8)), leading to either very long conversion times or
disproportionate amounts of the required enzyme when convert-
ing this sugar. In addition, conversions at elevated temperatures
are desirable for applications in industry, as catalytic activities
increase with higher temperatures and undesired microbiological
growth is avoided as well. A thermostable variant of P2Ox with
Spadiut et al.
D
the P2Ox-encoding insert was sequenced using primers T7promfwd
(50-AATACGACTCACTATAGGG-30) and T7termrev (50-GCTAGT-
TATTGCTCAGCGG-30).
Protein Expression and Purification. P2Ox protein was expressed in
E. coli BL21 Star DE3 cells and purified by immobilized metal affinity
chromatography according to previously published protocols (13,14). The
eluted, concentrated enzymes were washed 3 times using 10 mL of
KH2PO4 buffer (50 mM, pH 6.5), and finally diluted in the same buffer
to a protein concentration of 10-20 mg/mL.
Enzyme Activity Assays. P2Ox activity was measured with the
standard chromogenic ABTS [2,20-azinobis(3-ethylbenzthiazolinesulfonic
acid)] assay (2). A sample of diluted enzyme (10 μL) was added to 980 μL
of assay buffer containing horseradish peroxidase (142 U), ABTS
(14.7 mg) and KH2PO4 buffer (50 mM, pH 6.5). The reaction was started
increased catalytic activity with
D-galactose would therefore be of
applied interest for food industry.
In this study, we examined various active-site variants of P2Ox
isolated from the white rot fungus Trametes multicolor (TmP2Ox)
for increased activity with D-galactose. Furthermore, we com-
bined various mutations, both at the active site and the subunit
interface and known from former studies on P2Ox (13-15), to
create a tailor-made, stable biocatalyst for the conversion of
hydrolyzed lactose.
by adding D-glucose (20 mM). The absorbance change at 420 nm was
recorded at 30 °C for 180 s. The molar absorbance coefficient at 420 nm
(ε420) used was 42.3 mM-1 cm-1. One unit of P2Ox activity was defined as
3
the amount of enzyme necessary for the oxidation of 2 μmol of ABTS per
min, which equals the consumption of 1 μmol of O2 per min, under assay
conditions. Protein concentrations were determined by the Bradford assay
using the BioRad Protein Assay Kit with bovine serum albumin as
standard. All samples were measured at least in duplicates, and errors of
these repeated measurements were always below 5%.
Steady-State Kinetic Measurements. Kinetic constants were calcu-
lated by nonlinear least-squares regression, fitting the data to the Henri-
Michaelis-Menten equation. These constants were measured for the two
MATERIALS AND METHODS
Plasmids, Microorganisms and Media. The construction of the
pET21dþ/P2Ox vector (pHL2), which expresses the C-terminally His-
tagged P2Ox gene from Trametes multicolor (GenBank Accession No.
AY291124) under control of the T7 promoter, has previously been
described (16). The plasmids carrying the V546C as well as the E542K
mutations have also been described elsewhere (13, 14). Active, recombi-
nant wild-type TmP2Ox and mutational variants of P2Ox were expressed
in Escherichia coli strain BL21 Star DE3 (Invitrogen; Carlsbad, CA).
TBamp-medium was used to cultivate E. coli cells for protein expression
under appropriate selective conditions (ampicillin was added to 0.1 g/L).
All chemicals used were purchased from Sigma (Vienna, Austria) and were
of the highest grade available.
Generation of Mutants and Screening for Improved Variants.
Based on the crystal structure (PDB 1TT0 (17)), position His450 of
TmP2Ox was targeted by site-saturation mutagenesis, which allowed the
construction of a P2Ox variant library containing all possible codons at
this position. The library size was determined by the mutagenic codon
NNN (18). A total number of 360 colonies rather than the required 190
were screened to statistically cover more than 95% of all possible
combinations with high probability. A convenient 96-well plate-screening
assay, which allows a fast and reliable selection of improved P2Ox
variants, has been described earlier (13). In brief, a total of six wells per
plate were inoculated with E. coli carrying the wild-type P2Ox gene as a
control. The average value of the specific activity of recombinant wild-type
P2Ox (His6-rP2Ox) with the substrate of interest was determined, and
compared with activities for the P2Ox variants. The screening assay was
electron donors
D
-glucose (0.1-50 mM) and -galactose (0.1-200 mM)
using the standard ABTS assay and oxygen (air saturation).
D
Thermal Stability. Kinetic stability of the TmP2Ox variants was
determined by incubating the enzymes in appropriate dilutions in 50 mM
phosphate buffer (pH 6.5) at 60 and 70 °C and by subsequent measure-
ments of the enzyme activity (A) at various time points (t) using the
standard ABTS assay and glucose as the substrate (20 mM). A thermal
€
cycler (thermocycler T3, Biometra; Gottingen, Germany) and thin-walled
PCR tubes were used for all thermostability measurements. Residual
activities (At/A0, where At is the activity measured at time t and A0 is the
initial P2Ox activity) were plotted versus the incubation time. The
inactivation constant kin was obtained by linear regression of (ln activity)
versus time. The half-life values of thermal inactivation τ1/2 were calculated
using τ1/2 = ln 2/kin (20). At room temperature no inactivation of the
enzyme in this buffered system was observed within the 48 h.
Electrophoresis. Electrophoresis was performed principally as des-
cribed by Laemmli (21) to check the purity and the correct molecular mass
of the purified mutational variants. Native PAGE was performed using
a 5% stacking gel and a 10% separating gel in a PerfectBlue vertical
electrophoresis system (Peqlab; Erlangen, Germany). Samples were di-
luted to 1-2 mg of protein per mL, and aliquots of 5 μL were loaded per
lane. The High Molecular Weight Calibration Kit (Amersham; Piscat-
away, NJ) was used as mass standard. Staining was performed with
Coomassie blue.
performed with the two sugar substrates
vigorous shaking to achieve saturating oxygen levels. Variants showing
concomitantly high catalytic activity with -galactose and reduced activity
with -glucose compared to the recombinant wild-type enzyme were
D-glucose and D-galactose using
D
D
selected, and their genes were sequenced to determine the introduced
mutation at position His450.
Bioreactor Cultivations. To produce sufficient amounts of the
recombinant wild-type enzyme and the mutational variants H450G/
V546C and H450G/E542K/V546C, batch cultivations were carried out
in a 5 L Biostat MD stirred tank reactor (Braun; Melsungen, Germany)
For the introduction of site-directed mutations, the P2Ox gene was
mutated by a two-step mutagenesis method using PCR and digestion with
DpnI (19). The plasmid carrying the P2Ox gene was used as a template for
site-directed mutagenesis at position His450 with the primers H450G_fwd
(50-CACTCAGATCGGGCGCGATGCTTTCAGTTACGG-30) and
H450G_rev (50-CCGTAACTGAAAGCATCGCGCCCGATCTGAG-
TGT-30). The additional mutations V546C and E542K were introduced
as previously reported (13-15). The PCR reaction mix contained 2.5 U
Pfu DNA polymerase (Fermentas; St. Leon-Rot, Germany), 100 ng of
plasmid DNA, 5 pmol of each primer, 10 μM of each dNTP and 1ꢀ PCR
buffer (Fermentas) in a total volume of 50 μL. The mutagenic PCRs were
done using the following conditions: 95 °C for 4 min, then 30 cycles of
94 °C for 30 s; 58 °C for 30 s; 72 °C for 16 min, and a final incubation at
72 °C for 10 min. After PCR, the methylated template DNA was
degraded by digestion with 10 U of DpnI at 37 °C for 3 h. The remaining
PCR products were separated by agarose gel electrophoresis and puri-
fied using the Wizard SV Gel and PCR-Clean-Up System (Promega;
Madison, WI). Five microliters of the PCR products were transformed
into electrocompetent E. coli BL21 Star DE3 cells. To confirm the
presence of the correct mutations and the absence of undesired mutations,
with a working volume of 4 L. These cultivations were done in TBamp
-
medium. A preculture grown in an Erlenmeyer flask containing 200 mL of
TBamp-medium (37 °C, 140 rpm) was transferred to the bioreactor at an
OD600 of ∼0.5. The temperature was set to 25 °C, the culture pH was
maintained at pH 7.0 by automatic addition of sterile NaOH (4 M), and the
dissolved oxygen concentration (DO2) was set to 30%. The DO2 level was
maintained by automatically supplying filtered air (0-4 L/min) and
adjusting the stirrer velocity. All parameters were controlled by a digital
control unit (IMCS-2000, PCS AG; Wetzikon, Switzerland). The medium
was supplemented with 5 g/L lactose for induction of P2Ox expression
from thebeginningofthe cultivation. Samples were taken every 2 h, andthe
optical density, total intracellular protein concentration and P2Ox activity
were measured to monitor recombinant enzyme production. When the
volumetric activity reached a maximum, cells were harvested by centrifuga-
tion and homogenized and enzymes were purified as described before (13).
Batch Conversion Experiments. His6-rP2Ox and the variants
H450G/V546C and H450G/E542K/V546C were compared in terms of

Article
J. Agric. Food Chem., Vol. 58, No. 6, 2010 3467
their ability to concomitantly oxidize
D
-glucose and
D
-galactose to the
selected variants, the amino acid glycine was encoded by either of
the four possible codons (GGG, GGA, GGC, GGT), which
proved the success of saturation mutagenesis and the reliability of
the screening assay. In order to further improve this variant for
the possible conversion of hydrolyzed lactose in food industry, we
introduced additional mutations. Variant V546C is characterized
corresponding 2-ketoaldoses using oxygen as electron acceptor. Six batch
conversion experiments (each with a volume of 300 mL) were performed in
parallel in a multifermenter (Infors; Bottmingen, Switzerland). The
specific activities of the enzyme samples used were 6.3 U/mg for His6-
rP2Ox, 2.7 U/mg for H450G/V546C, and 1.3 U/mg for H450G/E542K/
V546C, employing standard assay conditions and
strate. In order to compare these three variants directly in the disconti-
nuous conversions, we used a total of 25 -glucose units of each enzyme for
the corresponding experiments. Catalase was used in excess (100,000 U)
to decompose H2O2. The conversion experiments were conducted in
100 mM KH2PO4 buffer (pH 6.5) at 30 °C and at 50 °C, 300 rpm and a
DO2 concentration of 20%. Conversion reactions were performed with
D-glucose as the sub-
by elevated kcat values for both sugar substrates
-galactose, which is, however, accompanied by elevated KM
values (14). Variant E542K is characterized by reduced KM values
for both -glucose and -galactose and significantly increased
D-glucose and
D
D
D
D
stability (13). His-tagged recombinant wild-type P2Ox (His6-
rP2Ox) and the variants H450G, H450G/V546C and H450G/
E542K/V546C were expressed and purified to apparent homo-
geneity (Figure 1) for further characterization.
Kinetic Characterization of Mutational Variants. Initial rates of
substrate turnover were recorded over a substrate range of 0.1 to
100 mmol/L of each sugar substrate D-glucose and D-galactose. Samples
(2 mL) were taken periodically, held at 95 °C for 3 min to inactivate the
enzymes and clarified by centrifugation. The supernatants were analyzed
for their sugar content using high performance anion exchange chroma-
tography with pulsed amperometric detection (HPAEC-PAD), which was
carried out using a Dionex DX-500 system (Dionex; Sunnyvale, CA) and a
CarboPac PA-1 column (4 ꢀ 250 mm) at 30 °C (22).
50 mM D-glucose and 0.1 to 200 mM D-galactose for His6-rP2Ox
and the mutational variants using the standard ABTS assay and
air oxygen. Kinetic data are summarized in Table 1. Variant
H450G showed similar binding (Michaelis constant KM) of
RESULTS
D
-glucose as His6-rP2Ox, but a 3.8-fold reduction in the turnover
number kcat, overall resulting in a 4-fold decreased catalytic
efficiency kcat/KM. On the contrary, kcat/KM for -galactose was
increased 5-fold because of a significant reduction in KM and
slightly increased kcat value for this substrate. Variant V546C of
TmP2Ox has been described earlier (14), and is characterized by
dramatically increased KM and kcat values for both sugar sub-
strates. The double mutant H450G/V546C showed elevated KM
Generation of Mutants. The strictly conserved active-site
loop 450HRDAFSYGA459 of pyranose oxidase is highly dynamic
and has been proposed to play an important role in sugar subs-
trate binding and discrimination (16,23). In order to improve the
reactivity of TmP2Ox with various substrates we chose the
residues in the active-site loop as targets for saturation mutagen-
D
esis, aiming at improved reactivity with the substrate D-galactose.
His450 is found at a pivotal position at the base of this loop (23),
and was one of the targeted residues selected for mutagenesis
studies (24). A screening assay based on microtiter plates (13)
revealed several His450 variants with improved catalytic acti-
values for both sugar substrates
compared to His6-rP2Ox (2.6-fold and 1.4-fold, respectively).
The turnover number for -glucose decreased significantly (2.9-
fold), whereas it increased 2.4-fold for -galactose resulting in a
1.7-fold increase in catalytic efficiency for -galactose, while kcat
KM for -glucose was reduced 3.8-fold. The mutation E542K in
TmP2Ox results in a decrease of both KM and kcat value for the
sugar substrates -glucose and -galactose, as has been shown
D-glucose and D-galactose
D
D
vity for
D-galactose and concomitantly reduced activity with
D
/
D-glucose, all of which were identified as His450Gly. In these
D
D
D
before (13). This was confirmed when introducing this muta-
tion into variant H450G/V546C. Overall, this triple mutant
H450G/E542K/V546C showed a catalytic efficiency for its sub-
strate D-galactose that was increased 1.2-fold compared to His6-
rP2Ox, while kcat/KM for
D-glucose was decreased to 17% of its
original value.
His6-rP2Ox clearly prefers
D
-glucose to -galactose as its sugar
D
substrate; this is also expressed by the substrate selectivity value,
i.e., the ratio of the catalytic efficiencies kcat/KM for the two
substrates (25). This value is 177 for His6-rP2Ox, while it is 8.9
for H450G, 27.4 for H450G/V546C and 24.4 for H450G/E542K/
V546C (Table 1), indicating that these enzyme variants do not
Figure 1. Native PAGE analysis of Trametes multicolor His-tagged
recombinant wild-type P2Ox (His6-rP2Ox) and mutational variants
H450G/V546C and H450G/E542K/V546C. Lane 1, molecular mass mar-
ker proteins; lanes 2, 3 and 4, His6-rP2Ox, H450G/V546C and H450G/
E542K/V546C after purification by IMAC, respectively.
show such a strong preference for
more.
D-glucose as substrate any-
Table 1. Kinetic Properties and Substrate Selectivity of His-Tagged Recombinant Wild-Type P2Ox from Trametes multicolor and Mutational Variants with Either
D-Glucose or D-Galactose as Substrate and O2 (Air) as Electron Acceptora
D-glucose
D-galactose
kcat/KM
(mM-1 s-1
kcat/KM
(mM-1 s-1
selectivity:
(kcat/KM)Glc/(kcat/KM)Gal
enzyme
His6-rP2Ox
H450G
K
M (mM)
k
cat (s-1
)
)
K
M (mM)
k
cat (s-1
)
)
3
3
0.939 ( 0.04
0.987 ( 0.05
3.06 ( 0.14
2.43 ( 0.14
0.521 ( 0.02
0.770 ( 0.17
48.1 ( 0.50
12.5 ( 0.15
88.6 ( 1.30
16.8 ( 0.27
35.9 ( 0.33
6.81 ( 0.35
51.2
12.7
29.0
13.5
68.9
8.83
8.79 ( 0.54
2.45 ( 0.12
46.2 ( 3.21
12.0 ( 0.38
3.87 ( 0.30
7.80 ( 0.70
2.51 ( 0.05
3.51 ( 0.04
6.57 ( 0.15
5.92 ( 0.05
2.59 ( 0.04
2.84 ( 0.06
0.286
1.43
177
8.88
210
27.4
103
24.4
V546Cb
0.142
0.493
0.669
0.364
H450G/V546C
E542Kc
H450G/E542K/V546C
a Substrate selectivity is given as the ratio of the catalytic efficiencies kcat/KM for the two substrates D-glucose and D-galactose. b Data from Spadiut et al. (14). c Data from
Spadiut et al. (13)

3468 J. Agric. Food Chem., Vol. 58, No. 6, 2010
Spadiut et al.
Table 2. Kinetic Stability of His-Tagged Recombinant Wild-Type P2Ox from T. multicolor and Mutational Variants at 60 °C and 70 °C
60 °C
inactivation constant kin (min-1
70 °C
inactivation constant kin (min-1
variant
His6-rP2Ox
H450G/V546C
)
half-life τ1/2 (min)
)
half-life τ1/2 (min)
687 ꢀ 10-4
345 ꢀ 10-4
nd
10
nda
<1
<1
62
20
>1200
nd
H450G/E542K/V546C
112 ꢀ 10-4
a Not determined.
enzyme was used for the corresponding conversion experiments,
which were conducted at both 30 and 50 °C (Figure 3, Table 3). At
30 °C His6-rP2Ox clearly preferred
D-glucose to D-galactose as
D-glucose was completely consumed,
substrate. Only when
-galactose was converted as evident from HPLC analysis, albeit
at a very low rate of 0.018 g L-1 h-1 during the batch conversion
D
3
3
(and hence is not evident in Figure 3A because of the scale used).
This conversion rate was 130-fold lower than the value of
2.34 g L-1 h-1 calculated for
D
-glucose, again stressing the
-galactose oxidation versus that
-glucose. At 50 °C the clear preference of the rec wt enzyme
-glucose was again obvious. Because of thermal inactivation
3
3
prominent discrimination of
of
for
D
D
D
over time, the conversion rate rapidly dropped within approxi-
mately 10 h, and the enzyme was completely inactivated after 15 h
as was evident from residual
the complete lack of 2-keto-
tion of -glucose and -galactose when using His6-rP2Ox was
D-glucose in the reaction mixture and
D-galactose. This sequential oxida-
D
D
shown before, both when using oxygen and benzoquinone as
electron acceptor during the bioconversions (14).
Variant H450G/V546C converted both sugar substrates
concomitantly and showed a 4-fold improved conversion of
Figure 2. Inactivation kinetics of pyranose oxidase from T. multicolor, both
recombinant His-tagged wild-type enzyme and variants, showing the
residual activity after incubation at increased temperature for various time.
Symbols: [, His6-rP2Ox at 60 °C; 9, variant H450G/V546C at 60 °C; ꢀ ,
variant H450G/E542K/V546C at 70 °C; 2, variant H450G/E542K/V546C
at 60 °C.
D-galactose compared to His6-rP2Ox at 30 °C. Similar as for
His6-rP2Ox, the conversion stopped after approximately 15 h
at 50 °C, presumably again because of thermal inactivation of
the enzyme. Because of the reduced conversion rate of
D
-glucose, ∼25% of the initial amount of
detected in the reaction mixture at that time.
Variant H450G/E542K/V546C showed the highest conversion
rate of -galactose in the batch conversion studies. This triple
mutant converted both sugar substrates concomitantly at
high rates of 1.78 g L-1 h-1 for -glucose and 0.23 g L-1 h-1
D-glucose was still
Thermal Stability. Kinetic stability, defined as the length of
time in which an enzyme remains active before undergoing
irreversible inactivation, of His6-rP2Ox and of variants H450G/
V546C and H450G/E542K/V546C was determined at 60 °C
and at 70 °C and a constant pH of 6.5. The inactivation cons-
tants kin and the half-lives of denaturation τ1/2 were determined
(Table 2). All enzymes showed first-order inactivation kinetics
when analyzed in the ln(residual activity) versus time plot
(Figure 2). The mutation E542K in combination with H450G
and V546C stabilized P2Ox significantly. At 60 °C the half-life
was increased by more than 120-fold compared to the rec
wt enzyme and the double mutant. The effect of the muta-
tion on stability was also obvious at 70 °C, where H450G/
E542K/V546C was still active after 2 h with a τ1/2 of 62 min,
whereas His6-rP2Ox and H450G/V546C inactivated comple-
tely within 1 min.
D
D
3
3
3
3
for
-galactose was observed for this thermostable variant at 50 °C,
reaching a value of 0.29 g L-1 h-1; this corresponds to a 16-fold
D-galactose. A further increase in the conversion rate for
D
3
3
increase compared to His6-rP2Ox employed at 30 °C.
DISCUSSION
Lactose is the main constituent of milk solids, accounting for
40% of the solids in cow’s milk. β-D-Galactosidase (EC 3.2.1.23)
efficiently hydrolyses lactose into
D
-glucose and -galactose (11),
D
and hence these sugars can be obtained in huge amounts as
lactose hydrolysates. They have significant potential as a fermen-
tation substrate, but can also be used as substrates for further
enzymatic transformations to more valuable food ingredients.
A conceivable transformation is the isomerization of these
Enzyme Production and Substrate Conversion Experiments.
Batch cultivations of E. coli BL21DE3 carrying the genes for
either the His-tagged wild-type or the mutated enzymes H450G/
V546C and H450G/E542K/V546C were performed to obtain
sufficient amounts of protein for the following conversion experi-
ments. Biomass was harvested at an OD600 of ∼8.0, when the
D-glucose/D-galactose mixtures to the corresponding ketoses
D-fructose and
D
-tagatose. -Tagatose is almost as sweet as
D
sucrose (92% relative sweetness in 10% solutions) yet its caloric
value is estimated to be almost negligible (26), and therefore its
use allows a significant reduction in caloric value of food at a
volumetric activity of His6-rP2Ox with
D-glucose reached a
maximum level of 230 U L-1, corresponding to approximately
comparable level of sweetness. Both D-fructose and D-tagatose
3
30 mg of recombinant protein per liter. The purified enzymes were
used for batch conversion experiments with an equimolar mixture
can be obtained from the respective aldoses by employing
isomerases, glucose (xylose) isomerase and arabinose isomerase,
respectively (26). However, only the equilibrium concentrations
can be attained when using this approach. Another biocatalytic
of
D-glucose and
D-galactose (100 mM each) as substrate and
oxygen as electron acceptor. Twenty-five
D-glucose units of each

Article
J. Agric. Food Chem., Vol. 58, No. 6, 2010 3469
Figure 3. Batch conversion experiments of an equimolar mixture of -glucose and -galactose (100 mM each) and oxygen as electron acceptor in saturating
concentrations, using His-tagged recombinant wild-type TmP2Ox, the variant H450G/V546C or H450G/E542K/V546C as biocatalysts. A, His6-rP2Ox and
oxygen at 30 °C; B, His6-rP2Ox and oxygen at 50 °C; C, H450G/V546C and oxygen at 30 °C; D, H450G/V546C and oxygen at 50 °C; E, H450G/E542K/
D
D
V546C and oxygen at 30 °C; F, H450G/E542K/V546C and oxygen at 50 °C. Symbols: b, D-glucose; 9, D-galactose.
Table 3. Batch Conversion Experiments of Equimolar Mixtures of D-Glucose and D-Galactose (100 mM Each) Performed at 30 °C and at 50 °C and Using His-
Tagged Recombinant Wild-Type Pyranose Oxidase from T. multicolor and the Variants H450G/V546C and H450G/E542K/V546Ca
batch A
batch B
batch C
batch D
batch E
batch F
enzyme
temp (°C)
His6-rP2Ox
30
2.34
0.018b
His6-rP2Ox
50
2.30
H450G/V546C
30
H450G/V546C
50
H450G/E542K/V546C
30
H450G/E542K/V546C
50
D-Glc conversion (g L-1 h-1
)
1.49
0.068
1.18
0.023
1.78
0.23
2.10
0.29
3
3
D-Gal conversion (g L-1 h-1
)
0.00
3
3
b
a These biocatalysts were added based on equal P2Ox activities with D-glucose under standard assay conditions. D-Galactose was not converted until D-glucose was
completely oxidized as was evident from HPAEC analysis.
approach is via the corresponding 2-ketosugars 2-keto-
D
-glucose
P2Ox with
(5.7% relative activity). This simultaneous conversion of
and -galactose is important when lactose hydrolysates are used
as a starting material for the envisaged bioconversion, since P2Ox
is known to overoxidize its primary reaction product, 2-keto-
D
-galactose is too low relative to that of
D
-glucose
and 2-keto- -galactose, which subsequently can be easily and
D
D-glucose
efficiently reduced by catalytic hydrogenation to yield the corre-
sponding ketoses (10, 27). As the P2Ox-catalyzed reaction is
quasi-irreversible (28), close to theoretical yields can be obtained.
D
For the simultaneous conversion of
D-glucose and
D-galactose in
D-glucose, thus forming 2,3-diketo-D-glucose (9), once the pri-
biotechnological processes, however, the catalytic activity of
mary oxidation at C-2 is completed.

3470 J. Agric. Food Chem., Vol. 58, No. 6, 2010
Spadiut et al.
In several previous studies the improvement of TmP2Ox both
in terms of stability and reactivity was discussed (13, 14).
Recently, we reported on the stable P2Ox variant T169G/
In addition, this work shows that individual mutations identified
in TmP2Ox can be combined beneficially, enabling the design of
biocatalyts with desired and tailored properties as our knowledge
on structure/function relationship on P2Ox increases.
V546C, which converts
at identical rates (15). However, the conversion rates were
rather low (0.05 g L-1 h-1 for -glucose and 0.06 g L-1 h-1
D-glucose and D-galactose concomitantly
D
ABBREVIATIONS USED
3
3
3
3
for D-galactose, respectively), making this variant less interesting
P2Ox, pyranose 2-oxidase; FAD, flavin adenine dinucleotide;
TmP2Ox, pyranose 2-oxidase from Trametes multicolor; His6-
rP2Ox, His-tagged recombinant wild-type pyranose 2-oxidase;
ABTS, 2,20-azinobis(3-ethylbenzthiazolinesulfonic acid); PAGE,
polyacrylamide gel electrophoresis; DO2, dissolved oxygen con-
centration;
for industry because of very long process times.
In order to improve P2Ox with respect to its substrate
promiscuity and selectivity (i.e., increased activity with its poor
substrate D-galactose), we used a semirational approach, namely,
saturation mutagenesis of residue His450, which is located at the
active site loop involved in substrate recognition (23). When
screening the libraries of P2Ox variants mutated at position 450,
LITERATURE CITED
we found several mutants that reacted faster with
D-galactose and
(1) Volc, J.; Denisova, N. P.; Nerud, F.; Musılek, V. Glucose-2-oxidase
´
concomitantly slower with -glucose than His6-rP2Ox. All of
D
activity in mycelia cultures of basidiomycetes. Folia Microbiol. 1985,
30, 141–147.
(2) Danneel, H.-J.; Ullrich, M.; Giffhorn, F. Goal-oriented screening
method for carbohydrate oxidases produced by filamentous fungi.
Enzyme Microb. Technol. 1992, 14, 898–903.
these variants were identified as His450Gly. The replacement of
the aromatic, polar amino acid histidine by the small, apolar
glycine apparently influenced the positioning and/or flexibility of
the active site loop, which drastically affected the catalytic
characteristics of this enzyme variant. Variant H450G has been
crystallized, is currently analyzed in more detail in our laboratory
and will be discussed elsewhere (unpublished data).
A mutation in TmP2Ox that was recently identified by us is
V546C, which is characterized by significant increases in kcat for
both D-galactose and D-glucose, albeit at the cost of elevated
Michaelis constants (14). Since increased kcat values are desirable
for technological processes, where substrates are converted at
high concentrations well above KM, we combined this mutation
with H450G, resulting in the double mutant H450G/V546C. This
double mutant indeed showed increased kcat values, but also
higher KM values than H450G. In order to create a thermostable
(3) Leitner, C.; Haltrich, D.; Nidetzky, B.; Prillinger, H.; Kulbe, K. D.
Production of
a novel pyranose 2-oxidase by basidiomycete
Trametes multicolor. Appl. Biochem. Biotechnol. 1998, 70-72, 237–
248.
(4) Tasca, F.; Timur, S.; Ludwig, R.; Haltrich, D.; Volc, J.; Antiochia,
R.; Gorton, L. Amperometric biosensors for detection of sugars
based on the electrical wiring of different pyranose oxidases and
pyranose dehydrogenases with osmium redox polymers on graphite
electrodes. Electroanalysis 2007, 19, 294–302.
(5) Yamanouchi, T.; Ogata, N.; Tagaya, T. Clinical usefulness of serum
1,5-anhydroglucitol in monitoring glycaemic control. Lancet 1996,
347, 1514–1518.
(6) Neidleman, S. L., Amon, W. F., Geigert, J. Process for the produc-
tion of fructose. US Patent 4,246,347 to Cetus Corporation, 1981.
variant of P2Ox, which converts
D-glucose and D-galactose
concomitantly at high rates, we combined H450G/V546C with
E542K, which had shown positive effects on the catalytic proper-
ties and stability of TmP2Ox before (13, 14). The resulting triple
mutant H450G/E542K/V546C shows properties that make it
interesting for applications in food industry. Recombinant wild-
type P2Ox clearly prefers D-glucose to D-galactose resulting in a
substrate selectivity, the ratio of kcat/KM for these two substrates,
of 177. This ratio changes to 24.4 for H450G/E542K/V546C, due
€
(7) Freimund, S.; Huwig, A.; Giffhorn, F.; Kopper, S. Rare keto-aldoses
from enzymatic oxidation: substrates and oxidation products of
pyranose 2-oxidase. Chem.;Eur. J. 1998, 4, 2442–2455.
(8) Leitner, C.; Volc, J.; Haltrich, D. Purification and characterization
of pyranose oxidase from the white-rot fungus Trametes multicolor.
Appl. Environ. Microbiol. 2001, 67, 3636–3644.
(9) Giffhorn, F. Fungal pyranose oxidases: occurrence, properties and
biotechnical applications in carbohydrate chemistry. Appl. Micro-
biol. Biotechnol. 2000, 54, 727–740.
(10) Haltrich, D.; Leitner, C.; Neuhauser, W.; Nidetzky, B.; Kulbe,
K. D.; Volc, J. A convenient enzymatic procedure for the production
of aldose-free D-tagatose. Anal. N.Y. Acad. Sci. 1998, 864, 295–
299.
to an increase in kcat/KM for
D-galactose and a decrease in kcat/KM
for -glucose. In contrast to the previously reported variant
D
T169G/V546C (15), the kcat value for
and also was reasonably high for
His6-rP2Ox. While His6-rP2Ox converted
D
D
-galactose was increased
-glucose compared to
-galactose only when
-glucose was depleted from the reaction mixture, H450G/
D
€
(11) Nguyen, T.-H.; Splechtna, B.; Steinbock, M.; Kneifel, W.; Lettner,
D
H. P.; Kulbe, K. D.; Haltrich, D. Purification and characterization of
two novel β-galactosidases from Lactobacillus reuteri. J. Agric. Food
Chem. 2006, 54, 4989–4998.
E542K/V546C catalyzes the concomitant oxidation of both
sugars, as was confirmed in small-scale bioconversion experi-
ments. Introducing the E542K mutation into the variant enabled
conversions at higher temperatures, which is preferable because
of higher reaction rates and a decreased possibility of micro-
bial contamination. The triple mutant showed considerably
increased thermostability as is evident from a remarkable increase
in half-life times, both at 60 and 70 °C. Thus, bioconversions
based on this thermostable variant will be feasible at temperatures
of up to 60 °C.
In conclusion, the semirational approach selected for the
engineering of P2Ox with respect to its substrate specificity and
promiscuity proved successful. By combining the newly identified
H450G variant with other mutations we obtained a thermostable
biocatalyst that shows significantly improved catalytic properties
compared to His6-rP2Ox. This variant holds promise for applica-
tions in food industry for the conversion of hydrolyzed lactose
toward a sugar mixture of considerably increased sweetness.
ꢀ
(12) Gonzalez-Siso, M. The biotechnological utilization of cheese whey: a
review. Bioresour. Technol. 1996, 57, 1–11.
(13) Spadiut, O.; Leitner, C.; Salaheddin, C.; Varga, B.; Vertessy, B.; Tan,
T.-C.; Divne, C.; Haltrich, D. Improving thermostability and
catalytic activity of pyranose 2-oxidase from Trametes multicolor
by rational and semi-rational design. FEBS J. 2009, 276, 776–792.
(14) Spadiut, O.; Pisanelli, I.; Maischberger, T.; Peterbauer, C.; Gorton,
L.; Chaiyen, P.; Haltrich, D. Engineering of pyranose 2-oxidase:
improvement for biofuel cell and food applications through semi-
rational protein design. J. Biotechnol. 2009, 5, 250–257.
(15) Spadiut, O.; Radakovits, K.; Pisanelli, I.; Salaheddin, C.; Yamabhai,
M.; Tan, T. C.; Divne, C.; Haltrich, D. A thermostable triple mutant
of pyranose 2-oxidase from Trametes multicolor with improved
properties for biotechnological applications. Biotechnol. J. 2009, 4,
525–534.
(16) Kujawa, M.; Ebner, H.; Leitner, C.; Hallberg, B. M.; Prongjit, M.;
Sucharitakul, J.; Ludwig, R.; Rudsander, U.; Peterbauer, C.;
Chaiyen, P.; et al. Structural basis for substrate binding and

Article
J. Agric. Food Chem., Vol. 58, No. 6, 2010 3471
regioselective oxidation of monosaccharides at C3 by pyranose
2-oxidase. J. Biol. Chem. 2006, 281, 35104–35115.
(17) Hallberg, B. M.; Leitner, C.; Haltrich, D.; Divne, C. Crystal
structure of the 270 kDa homotetrameric lignin-degrading enzyme
pyranose 2-oxidase. J. Mol. Biol. 2004, 3, 781–796.
(18) Georgescu, R.; Bandara, G.; Sun, L. Saturation Mutagenesis. In
Directed Evolution Library Creation: Methods and Protocols; Arnold,
F. H., Georgiou, G., Eds.; Springer; New York; 2003; pp 75-83.
(19) Li, S.; Wilkinson, M. F. Site-directed mutagenesis: a two-step
method using PCR and DpnI. Biotechniques 1997, 4, 588–590.
(20) Polizzi, K. M.; Bommarius, A. S.; Broering, J. M.; Chaparro-
Riggers, J. F. Stability of biocatalysts. Curr. Opin. Chem. Biol.
2007, 11, 220–225.
(24) Salaheddin, C.; Spadiut, O.; Ludwig, R.; Tan, T.-C.; Divne, C.;
Haltrich, D.; Peterbauer, C. Probing active-site residues of pyranose
2-oxidase from Trametes multicolor by semi-rational protein design.
Biotechnol. J. 2009, 4, 535–543.
(25) Morley, K. L.; Kazlauskas, R. J. Improving enzyme properties: when
are closer mutations better? Trends Biotechnol. 2005, 23, 231–237.
(26) Oh, D. K. Tagatose: properties, applications, and biotechnological
processes. Appl. Microbiol. Biotechnol. 2007, 76, 1–8.
(27) Leitner, C.; Neuhauser, W.; Volc, J.; Kulbe, K. D.; Nidetzky, B.;
Haltrich, D. The Cetus process revisited: A novel enzymatic alter-
native for the production of aldose-free D-fructose. Biocatal. Bio-
transform. 1998, 16, 365–382.
(28) Prongjit, M.; Sucharitakul, J.; Wongnate, T.; Haltrich, D.; Chaiyen,
P. Kinetic mechanism of pyranose 2-oxidase from Trametes multi-
color. Biochemistry 2009, 48, 4170–4180.
(21) Laemmli, U. K. Cleavage of structural proteins during the assembly
of the head of bacteriophage T4. Nature 1970, 227, 680–685.
€
(22) Splechtna, B.; Nguyen, T.-H.; Steinbock, M.; Kulbe, K. D.; Lorenz,
W.; Haltrich, D. Production of prebiotic galacto-oligosaccharides
from lactose using β-galactosidases from Lactobacillus reuteri.
J. Agric. Food Chem. 2006, 54, 4999–5006.
Received for review November 17, 2009. Revised manuscript received
January 27, 2010. Accepted January 30, 2010. Financial support from
the Austrian Science Fund (FWF, Translational Project L213-B11) is
gratefully acknowledged. TTN thanks the Austrian Academic
(23) Hallberg, B. M.; Leitner, C.; Haltrich, D.; Divne, C. Crystal
structure of the 270 kDa homotetrameric lignin-degrading enzyme
pyranose 2-oxidase. J. Mol. Biol. 2004, 341, 781–796.
€
Exchange Service OAD for a Technology Grant Southeast Asia.