30
Glucose Oxidase
A. J. Vroemen
DSM Food Specialties, Delft, The Netherlands
I. INTRODUCTION acid with simultaneous production of hydrogen perox-
ide. The decrease of the pH caused by the gluconic acid
Glucose oxidase (GOX) is produced predominantly by as well as the antibacterial effect of the hydrogen per-
fungi as part of an enzyme complex also containing oxide produced, are favorable for the fungus. In addi-
catalase. The best-studied microorganisms are species tion, A. niger is able to consume the gluconic acid
of Aspergillus (1 3) and Penicillium (4, 5). In produced, which is not the case for most of the com-
Aspergillus the enzymatic complex is located in cell peting microorganisms. Bacteria like Gluconobacter are
compartments called peroxisomes (6), which results also capable of converting glucose into gluconic acid,
in the need to open the fungal cell walls before the but these bacteria use a different mechanism.
enzyme can be released and purified. The enzyme com- Gluconobacter oxidans contains two types of glucose
plex of Penicillium is secreted into the extracellular dehydrogenases, one PQQ and one NADP dependent,
environment (7). Almost all GOX preparations avail- which convert glucose to gluconic acid without forma-
able on the market are produced by Aspergillus niger. tion of hydrogen peroxide. The PQQ-dependent
These preparations are either highly or somewhat or enzyme seems to be the most important (8). This oxi-
completely exempt of catalase (see Sec. II.A.1). This dation process yields energy for the bacteria con-
chapter will focus mainly on GOX of Aspergillus niger. cerned. It is notable that these bacteria are rather
The question can be raised why fungi like acid tolerant.
Aspergillus niger invest a high amount of nutrients
and energy in the synthesis of the enzymatic complex. II. REACTION CATALYZED
As we will see later, the catalyzed reaction of glucose to
gluconic acid does not yield energy for the microorgan- As substrates GOX uses b(D)-glucose and oxygen.
ism. Most probably the enzymatic complex gives the Two hydrogen atoms are first transferred from glucose
microorganiism an ecological advantage in the soil. In to the coenzyme FAD during formation of d-glucono-
nature, GOX is produced by the fungus only when lactone (9). Subsequently the enzyme transfers the two
glucose is present: It is an inducible enzyme. When hydrogen atoms directly to molecular oxygen during
glucose is liberated from, for instance, starch or cellu- formation of hydrogen peroxide. The d-gluconolactone
lose, the synthesis of the enzymatic complex starts and is hydrolyzed to gluconic acid spontaneously or by the
in the presence of air, glucose is oxidized to gluconic enzyme gluconolactonase.
Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved.
cglu cO
2
C6H12O6 + E-FAD�C6H10O6 + E-FADH2 (1)
vźkcat �E�
Km �cglu Km �cO
glu O2 2
glucose gluconolactone
The Km values as determined by Gibson et al. (11)
C6H10O6 +H2O ! C6H12O7 (2)
Km ź110 mM; Km ź0:48 mM
glu O2
gluconolactone gluconic acid
(both determined at 278C.)
E-FADH2 +O2 ! E-FAD + H2O2 (3)
Here the molarity of glucose is defined as total glucose,
not as b-glucose.
C6H12O6 +H2O+O2�C6H12O7 + H2O2 The value of 110 mM mentioned by Gibson seems
rather high. Other values mentioned in the literature
glucose gluconic acid hydrogen
are 37 mM (12), which is more in agreement with GOX
peroxide
from other microorganisms. The Penicillium amagasa-
kiense enzyme has a higher affinity for glucose: a Km
The fact that GOX uses only b-glucose as substrate
value of 5.2 mM is reported (13). The problem in the
does not mean that glucose can be only partially oxi-
determination of the Michaelis-Menten constant for
dized; during the process a-glucose spontaneously
glucose is that precautions must be taken to ensure
mutarotates to a-,b-glucose, making all the glucose
that the concentration of dissolved oxygen, which
available as substrate. GOX is rather specific for glu-
is consumed during the reaction, remains constant.
cose. There is < 1% activity on related sugars like
Technically it seems possible to maintain the dissolved
xylose and galactose. Also, the activity on other nat-
oxygen concentration constant during the reaction in
ural mono- and disaccharides is very low. In a systema-
for instance a fermenter, enabling the determination of
tic study Whitaker showed that in the deoxyglucoses,
correct velocity values at different glucose concentra-
replacement of the hydroxyl group by a hydrogen
tions. However, to the author s best knowledge such an
atom at successively the 6, 4, 2, 3 and 5 position
approach has never been published so far. Initial velo-
resulted in respectively 10%, 4%, 3%, 1% and 0%
city, vo, can also be used to obtain correct data.
remaining activity (10).
These Km values mentioned above show that the
The hydrogen peroxide formed causes the reaction
affinities of GOX for the two substrates are rather
to stop since it inactivates the GOX rapidly and irre-
poor. This is especially true for the substrate oxygen
versibly. In principle, hydrogen peroxide can be hydro-
when we compare the Km for oxygen with the solubility
lyzed by the enzyme catalase, or the active oxygen
of oxygen in water. At 1 bar, 258C, and in the presence
can be transferred spontaneously or enzymatically to
of air, the maximum solubility of oxygen is 6 mg/L or
another molecule.
0.2 mM, equaling < 50% of the Km. Even with pure
As mentioned before, the fungi concerned also pro-
oxygen the maximum solubility of 30 mg/L is only
duce catalase, which oxidizes/reduces the hydrogen
twice the Km. With glucose concentrations far below
peroxide produced into water and oxygen. The overall
the Km value, which is between 7 to 20 g/L for the
reaction of the enzyme complex becomes:
Aspergillus enzyme, the velocity of the reaction will
be low. As a consequence it can be concluded that
2C6H12O6 +2H2O +2O2�2C6H12O7
for the elimination of low quantities of glucose or oxy-
+ 2H2O2 (4)
gen from certain foods, high quantities of enzyme and/
2H2O2 ! 2H2O +O2 (5) or long incubation times are necessary even under opti-
mal conditions. The affinity of the enzyme catalase for
Overall reaction: its substrate hydrogen peroxide is extremely low: Km
2C6H12O6 +O2 ! C6H12O7 (6) =1.1 M or 37.4 g/L. This aspect is very important in
certain applications.
B. Determination of GOX Activity
III. ENZYME CHARACTERISTICS
A. Michaelis-Menten Constants From a theoretical point of view, the analysis of GOX
is rather complicated. The enzyme has two substrates:
The enzyme has two substrates: glucose and molecular glucose and oxygen. In principle high glucose concen-
oxygen. Simple Michaelis-Menten kinetics results in trations can be chosen, far above the Km, so that the
Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved.
velocity of the reaction is no longer dependent on the tionship between IU and Sarrett units is determined
glucose concentration. This is more difficult for the as 1 IU=1.12 SU (16, 18) or 1.1 SU(17). (See Annex
second substrate oxygen. To obtain oxygen concentra- 1 for a detailed description of a method with o-dia-
tions in solution far above the Km, very high oxygen nisidine.) Methods based on the oxidation of a chro-
pressures have to be chosen, which is not practical. The mogen by hydrogen peroxide are now generally used.
concentration of oxygen is below the Km if the reaction The relationship between IU and Sarrett units can be
is performed under normal conditions, and as a result different for glucose oxidation from different organ-
the velocity of the reaction is below the maximum isms. Finally, it should be mentioned that comparison
velocity. of units in literature is extremely difficult owing to
An extra complication is that the reaction consumes different incubation conditions, including saturation
oxygen. When there is insufficient supply to maintain levels of oxygen, and different unit definitions. The
the oxygen concentration at a saturation level, there best thing to do is to work with an internal standard
will not be a linear production of gluconic acid in of GOX of known activity.
time at a fixed enzyme concentration or a linear rela-
tionship between the amount of gluconic acid pro-
duced within a fixed time and amount of enzyme. C. Specific Activity
Such linear relationships are desirable for an analytical
method. Therefore, an analytical method for GOX is By measuring the oxygen consumption of a pure, cat-
preferably based on a system in which only very small alase-free GOX preparation of A. niger, Tsuge et al.
quantities of oxygen are consumed. A sensitive method (21) found a specific activity of 172 IU/mg protein at
is not realized when it is based on the consumption of 308C and pH 5.6. About the same value has been
glucose or the formation of gluconic acid. If the ana- reported by Hayashi and Nakamura (22). Several
lysis is based on the consumption of oxygen, the result research groups have purified the A. niger enzyme by
is influenced by the presence of catalase. In this type of ion-exchange chromatography and obtained pure pre-
method high quantities of GOX-free catalase have to parations with activities from 200 to 250 Sarrett units/
be added. This is the principle underlying a method mg protein, equaling 180 225 IU/mg protein. Pure cat-
described in 1953 by Scott (14) and in 1957 by alase-free preparations can be obtained from compa-
Underkofler (15). In a Warburg equipment the amount
of oxygen consumed under optimal conditions was
determined. The unit most frequently used is the
Annex 1 Analysis of Glucose Oxidase
Sarrett unit, which is defined as the amount of enzyme
which catalyzes the uptake of 10 L oxygen/ min at
The method used by Whittington (26) is mentioned here:
308C, pH 5.9, and a glucose concentration of 3%. Glucose solution A: 2 g/L in 0.2 M Tris phosphate buffer,
pH 7.0.
Working with a Warburg is rather laborious, and
o-Dianisidine solution B: 2 g/L in 0.2 M Tris phosphate
therefore Underkofler also describes a method using
buffer pH 7.0.
a sodium hydroxide solution neutralizing the gluconic
Horseradish peroxidase solution C: 60 units/mL in 0.2 M
acid produced.
Tris phosphate buffer pH 7.0.
A very sensitive and accurate method is based on
Glycerol.
the production of hydrogen peroxide. The amount of
Enzyme solution standard D: 10 IU glucose oxidase/mL.
hydrogen peroxide is determined indirectly by the oxi-
Hydrochloric acid solution E: 5 M.
dation of a chromogen catalyzed by a peroxidase.
All reagents Sigma quality.
Owing to the very high affinity of this peroxidase
Preincubate all solutions at least 10 min at 308C.
for hydrogen peroxide compared to catalase, the pre- Add to 1 mL A, 0.1 mL B, 0.1 mL C, and 0.8 mL glycerol.
sence of this last enzyme in GOX preparations does Mix carefully.
not influence the analytical result. As chromogen, o- Add 10 20 L of an enzyme sample containing 0 10 IU/
mL.
dianisidine, o-toluidine, and others are used, the
Incubate 30 min at 308C.
developed color is measured in a spectrophotometer,
Stop reaction by adding 2 mL of 5 M hydrochloric acid
and by using internal standards the amount of oxi-
solution. Mix carefully.
dized glucose can be determined. The activity is often
Oxidized o-dianisidine is measured in a spectrophotometer
expressed as IU (international units), defined as
at 525 nm.
micromolar glucose oxidized per min under optimal
Compare with internal standard of 0 10 IU/mL enzyme.
conditions. For the Aspergillus niger enzyme the rela-
Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved.
nies selling research chemicals, or by growing a trans- of two identical subunits with 1 molecule FAD per
genic yeast harboring a GOX gene. subunit as coenzyme. Because of this, the color of the
The specific activity of the purified Penicillium ama- oxidized protein is yellow with absorption maxima at
gasakiense enzyme is 60% higher than that of the A. 377 and 455 nm. Under anaerobic conditions in the
niger enzyme (22). presence of glucose, the enzyme molecule is reduced
and the color disappears. When oxygen is admitted to
D. Turnover Number the system the color reappears. The FAD is not cova-
lently linked and can easily be removed by acid, urea, or
Gibson et al. (11) found a turnover number of 16,200/ guanidine. But without these reagents, the FAD is
min. Whittington et al. (27) cloned the A. niger enzyme bound to the enzyme. The apoenzyme is not active
in yeast and found a value of 17,000 20,000 for the but activity is restored upon incubation with FAD (15).
yeast-derived enzyme. Vmax values of 235/sec for the GOX is a glycoprotein. The amount of carbohydrate
wild-type enzyme and 500 for the yeast-derived enzyme in different preparations from A. niger can vary from
are found, but it remains unclear whether these differ- 10% to 18% (22, 23). Mannose is by far the most
ences are really significant. important sugar with amounts of 70 80%. Galactose
contributes 5% and glucosamine 15% to the total
E. pH Activity Profile carbohydrate content. In the Penicillium enzyme Kalisz
(13) found 95 residues of mannose, 12 residues of glu-
The pH activity curves for both the Penicillium and cosamine, and five residues of galactose per molecule of
Aspergillus enzymes show a horizontal profile between enzyme, and a total carbohydrate content of 13%.
pH 4.5 and 7.5, with a sharp decline on both sides (10, In the A. niger enzyme the sugar molecules are N
20). There is almost no activity at < pH 3.0 or > pH and O linked. By selective elimination of the O-linked
8.5. GOX is stabilized by its substrate glucose (32), sugars Takegawa (12) showed that predominantly
explaining why in certain applications with a pH mannose monomers are linked to serine and threonine,
between 2.5 and 3.0 and with a high concentration of but the exact place in the molecule is not yet known.
glucose the enzyme is still active. Also at > pH 8.0 the The N-linked sugar moiety can be selectively removed
stability of the enzyme is improved by adding the by an enzyme from Flavobacterium, resulting in the
substrate. liberation of 30% of the total carbohydrates. The
partly deglycosylated enzyme has the same kinetic
F. Temperature Activity Profile and biochemical properties and resistance to proteases
as the native enzyme except that the native enzyme
In determination of the temperature activity profile of precipitates at higher concentrations of ammonium
GOX, some specific complications are encountered. sulfate or TCA, and the deglycosylated enzyme is less
GOX is sensitive to the hydrogen peroxide formed espe- stable in relation to pH and temperature. The same
cially at higher temperatures. This means that when a results were obtained with the Penicillium enzyme by
method is applied in which the hydrogen peroxide is not Kalisz et al. (13). In this case 95% of the carbohydrate
removed immediately and completely, the optimum was split off by the enzymatic treatment. Losses in pH
temperature and the maximum temperature will be and temperature stability were observed as well.
low as compared to a method where this is done prop- The enzyme from Phanerochaete chrysosporium (24)
erly (10). Moreover, long incubation times give a lower was found to be a flavoprotein with a molecular weight
optimum. In the analytical method using perosidase of 160 kDa and containing 2 mol FAD per mol pro-
and a chromogen, the hydrogen peroxide reacts imme- tein, but it does not seem to contain any carbohydrate
diately with the chromogen. This method gives an activ- and its specificity is lower. Although glucose is the
ity profile which is rather constant with time between main substrate, considerable activity is still found on
308C and 608C. Above 608C activity goes down slowly sorbose, xylose, and maltose.
with still 10% activity at 708C but no activity at 808C. The gene of the A. niger enzyme has been isolated
and the sequence analyzed by various research groups
G. Protein Properties with identical results (25 27). This sequence is now
available in data bases. The gene codes for a signal
The molecular weights of the enzymes from both peptide of 22 amino acids (1 22), followed by the
Aspergillus and Penicillium have been determined at 583 amino acid subunit itself ( 23 605). The molecular
between 140 and 160 kDa (19 21). The enzyme consists weight of the subunit plus signal peptide is calculated
Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved.
to be 65,638, and without signal peptide it is 63,250. strate-binding site explains the high specificity of GOX
With 16% sugars and two molecules of FAD the described above. Part of the entrance to the pocket is
total molecular weight can be calculated to be 152 at the interface to the second subunit and is formed by
kDa. When the gene is expressed in yeast, an enzyme is a 20-residue lid. The carbohydrate moiety attached to
obtained with a higher glycolysation level and an Asn89 at the top of this lid forms a link between the
improved thermostability. This is in line with the subunits of the dimer. In total, there are four N-glyco-
observed lower thermostability the deglycosylated sylation sites, with an extended carbohydrate moiety at
enzymes. The kinetic parameters of the yeast-derived Asn89. Starting from the 3D structure of the
enzyme and the original one are not significantly dif- Pencillium amagasakiense enzyme, it could be shown
ferent. The yeast expression system can be used to by mutation of key conserved active-site residues that
obtain GOX completely free of catalase. The trans- Arg516 is involved in the binding to the 3-OH group of
genic yeast must not been grown on glucose since the the glucose molecule (31). Replacement of this arginine
hydrogen peroxide formed will inactivate the enzyme by another amino acid lowers the affinity for the sub-
and stop growth. Instead it can be grown on sacchar- strate. Aromatic residues on other locations like 73,
ose, since this sugar is hydrolyzed by membrane-bound 418, and 430 are important for the correct orientation
invertase in the cell to fructose and glucose, which and maximum velocity of glucose oxidation.
are immediately metabolized without formation of glu-
conic acid and hydrogen peroxide.
IV. APPLICATIONS OF GOX IN FOOD
H. Genetic Aspects
Looking at the reaction catalyzed by the enzyme, the
A comparison of the mature GOX sequence with applications are linked to four different aspects:
sequences present in data bases shows 26% homology
1. Removal of glucose
with the alcohol oxidase from Hansenula polymorpha.
2. Removal of oxygen
The GOX s from Aspergillus and Penicillium show
3. Production of gluconic acid
66% identity and 79% similarity. The Penicillium
4. Production of hydrogen peroxide
enzyme consists of 587 amino acids. The two enzymes
are highly conserved in the FAD-binding and sub- A separate subject is the application of glucose oxidase
strate-binding domain, in their secondary structures as an analytical agent.
and in regions at the subunit interface. The highest
similarity with other oxidoreductases is observed in A. Removal of glucose (10)
the FAD-binding domain. Aryloxidase, a FAD-depen-
dent enzyme involved in lignin degradation, has been Glucose is a reducing sugar that can react with amino
cloned from Pleurotus eryngii. The enzyme is com- groups in, for instance, proteins, to form colored
posed of 593 amino acids, 27 of which form a signal Maillard components that are often undesirable in
peptide. It shows 33% sequence identity with GOX food products. Additions of GOX to the food system
from Aspergillus niger. The predicted secondary struc- in the presence of air or an oxygen donor like hydrogen
tures of the two enzymes are very similar (28). peroxide will result in the conversion of glucose to the
non-amine-reactive gluconic acid and thus prevent the
I. Three-Dimensional Structure formation of these Maillard components. The best
known application is the prevention of nonenzymatic
The three-dimensional structures of the two enzymes browning in egg white powder. A detailed description
of Aspergillus and Penicillium have been determined by of this application is given by Scott (10). Treatment
x-ray crystallography at 2.3 Angstrom resolution (29) can be done up to 508C and in the pH range from 4
�
and later improved to 1.9 A resolution (30). The FAD- to 7. Egg white is neutral at the time of laying, but the
binding domain is very similar to other FAD-binding pH rises quickly as carbon dioxide is lost, such that it is
proteins, and 11 amino acid residues from different generally close to 9 when the egg white is processed.
parts of the molecule are involved in this binding. The pH is brought below 7 by adding citric acid; addi-
The same is true for substrate binding. The substrate tion of hydrogen peroxide as oxygen donor is advan-
enters a deep pocket and is stabilized by 12 hydrogen tageous. Owing to the low affinity of GOX for glucose,
bonds and hydrophobic contacts to three aromatic high quantities of enzyme and/or long incubation times
residues and to FAD. A detailed analysis of this sub- are necessary to remove the glucose almost completely,
Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved.
making this application less competitive than other into glucose and galactose, with GOX and hydrogen
systems like addition of active yeast. Canadian peroxide (40). Owing to the pH drop, the milk coagu-
researchers have shown that GOX is also efficient in lates. In principle, the same can be done in cheese
the reduction of nonenzymatic browning in potato manufacture, where direct acidification is fairly com-
products like chips and French fries (33, 34). mon. However, in industrial practice, the addition of
However, complete removal of glucose is also not an acid or gluconolactone, which under the conditions
realized in this process. It is clear that in this type of of cheese manufacturing slowly hydrolyzes into glu-
application only glucose is converted to a nonreducing conic acid, is preferred.
component; other sugars, such as maltose and lactose, Until, now production of gluconic acid on an indus-
maintain their reducing characteristics. A solution can trial scale is done by fermentation with selected strains
be the application of an oxidase active on a broad of Aspergillus niger and Gluconobacter oxydans.
range of reducing sugars. Some enzyme candidates Several attempts have been made toward production
are under investigation now. of gluconic acid using an enzymatic system. Until
recently complete bioconversion was only obtained
B. Removal of Oxygen with glucose solutions of 10% and lower, which is
not interesting from an economical point of view.
During storage of food, oxygen can have a detrimental Beverini and Vroemen (41) have now shown that glu-
effect on quality. As an example, oxidation of unsatu- cose concentrations as high as 40 50% can be comple-
rated fatty acids can lead to rancidity of vegetable oils. tely converted in a fermenter at pH 5 6 and up to
Oxidation of colored components will change the color 308C, using low quantities of GOX rich in catalase in
of beverages or wine. In addition, oxygen influences a relatively short time. The high catalase content
the taste of beer in a negative way during storage. needed is necessary owing to the low affinity of this
The majority of foods contain certain quantitites of enzyme for hydrogen peroxide. Under the conditions
glucose. In closed systems, the quantity of oxygen to mentioned, the concentration of hydrogen peroxide
be removed, in order to prevent oxidation, is generally remains low such that the two enzymes GOX and cat-
rather low. Therefore GOX can be used to remove alase are not completely inactivated before the end of
oxygen. If necessary a small quantity of glucose is the bioconversion. The advantages of this enzymatic
added. An important aspect is the stability of the process are that no fermentation is necessary (steriliza-
enzyme under application conditions. Fruit juices and tion, nutrients, time), the yield is up to 100%, and the
wine have a very low pH of 2.5 3, conditions in which purification of the final product is much easier.
the enzyme is neither active nor stable. However, the
high glucose concentrations in fruit juices appear to
D. Production of Hydrogen Peroxide
have a stabilizing effect on the enzyme.
A large number of publications describe the positive
Hydrogen peroxide is a potent oxidant, which can be
effect on quality of adding GOX to food. A good
used as an active antimicrobial agent. Based on this
review is given by Scott (10). Other publications con-
property of hydrogen peroxide, several applications
cern: prevention of rancidity in oils, fats and fish (35
of GOX have been developed of which some are
37); and prevention of off-flavors and color changes in
applied on a large scale. For these applications the
fruit juices, fruit concentrates, white wine, and beer
GOX preparations have to be exempt or poor in cat-
(10, 38, 39). Despite all these positive results, no appli-
alase. Unfortunately, the GOX itself risks also being
cations of adding GOX to food has been realized on an
inactivated by the hydrogen peroxide formed.
industrial scale until now. The main reason seems to be
the existence of competing technologies to prevent oxi-
1. Toothpaste
dation; for example, flushing out of oxygen by carbon
dioxide or nitrogen in beverages, addition of antioxi- In the Netherlands, toothpaste containing enzymes has
dants as BHA, ascorbic acid, sulfite, etc. been developed (42). If sufficient glucose is present, the
GOX generates hydrogen peroxide, which kills plaque-
C. Production of Gluconic Acid forming bacteria in the mouth. To obtain sufficient
glucose, amyloglucosidase can be added, which
Milk can be directly acidified by adding glucose, GOX, together with the amylase present in saliva, hydrolyzes
and hydrogen peroxide as oxygen donor. A variant is starch into glucose. In addition, it is claimed that GOX
combining lactase, which hydrolyzes the milk sugar stimulates the lactoperoxidase system in the mouth.
Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved.
2. Milk and Milk Products combination of GOX and sulfhydryl oxidase. The lat-
ter enzyme catalyzes the selective oxidation of sulfhy-
GOX can generate hydrogen peroxide for the lactoper-
dryl groups to disulfides by oxygen.
oxidase (LPO) system naturally present in milk. This
2 RSH�O2�RSSR�H2O2 �7�
combined GOX-LPO system has been studied exten-
sively by a number of research groups to solve severe
This leads to interprotein disulfide bonds. The role of
contamination problems in milk and cheese produc-
GOX is not yet completely understood. It clearly par-
tion (43, 44). Especially in cheese made from raw non-
ticipates in the formation of disulfide bonds between
pasteurized milk, there is an urgent need for a natural
gluten proteins, but most probably it also participates
antibacterial system. Although a strong effect could be
in the formation of other bonds like oxidative gelation.
shown on a high number of pathogens, introduction
This is the coupling of two ferulic acid residues of
of the GOX-LPO system has not yet taken place. One
neighboring arabinoxylan chains by the hydrogen-
of the reasons is that it has not given an absolute
peroxide formed by glucose oxidase.
guarantee against all pathogens.
The second patent (47) concerns the synergistic
effect of GOX and hemicellulases, like xylanases. It is
3. Baking
believed that GOX makes stronger doughs, permitting
the addition of higher amounts of hemicellulases.
An important aspect of baking is the strength or weak-
Addition of such amounts of hemicellulases alone
ness of the dough. Flours with a low protein content
often results in softer and sometimes sticky doughs.
are characterized as weak, and the gluten is very exten-
Especially, the latter combination of GOX and hemi-
sible under stress but does not return to its original
cellulases has obtained a high market acceptance. This
dimensions when the stress is released. Bakers gener-
  Hemilox  preparation, produced and commercialized
ally prefer strong doughs because of their better rheo-
by DSM Baking Ingredients under license of Cultor,
logical and handling properties, which result in a better
permits production of clean-label breads of high qual-
form and texture of the final bread. Bakers have used
ity without addition of nonspecific oxidants or emulsi-
dough conditioners to strengthen the dough. These
fiers (48).
conditioners are mostly nonspecific oxidants like bro-
mates, iodates, and ascorbic acid. In North America
4. Glucose Oxidase as an Analytical Agent
and Western Europe, public opinion and legislation
are more and more opposed to the use of chemicals
GOX is the most widely employed enzyme as an ana-
in bread. Moreover, these nonspecific oxidants can
lytical agent, particularly for the determination of glu-
have a negative influence on the bread aroma. In the
cose in clinical laboratories, fermentation media, food,
United Kingdom bromates are no longer allowed since
feed, etc. Different systems have been developed using
1990, and in France no addition of chemicals is per-
glucose oxidase in soluble form, but also in immobi-
mitted in the production of traditional bread (pain a`
lized form as glucose electrodes or sticks for the deter-
tradition franc� aise). Therefore, a need exists to replace
mination of glucose in urine, as practiced by every
the nonspecific oxidants by a natural alternative.
physician. For a good detailed review see Raba and
Enzymes are judged as natural and in agreement with
Mottola (49).
  clean label.  Already in 1957, a patent appeared (45)
which describes the addition of GOX to flour to
improve the dough quality and the baking properties.
ACKNOWLEDGMENTS
As a particular advantage, the combination with ascor-
bic acid is mentioned. This is already an indication that
The author wants to thank his son Casper W. Vroemen
GOX added gives a better result than ascorbic acid or
of Wageningen University for valuable comments on
other nonspecific oxidants alone, which has been con-
the manuscript.
firmed in many tests. This, together with the fact that
in that era the acceptance of chemicals was still high,
resulted in a low market penetration of the addition of
REFERENCES
GOX to flour.
The big breakthrough came in the beginning of the
1. B Drews, H Smalla. Brantweinwirtschaft 22, 1969.
1990s, when researchers of the Finnish company
2. K Zetelaki, K Vas. Biotech Bioeng 10:45, 1968.
Cultor discovered the synergistic effect of combining
3. W Franke, L Mochel, K Haye. Arch J Mikrobiol
several enzymes. The first patent (46) concerns the 51:323, 1965.
Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved.
4. T Yoshimura, T Isemura. J Biochem Tokyo 1971 25. KR Frederick, S Chakraborty, et al. J Biol Chem
69:839. 265:3793, 1990.
4a. S Nakamura, S Hayashi. FEBS Lett 41:327, 1974. 26. H Whittington et al. Curr Genet 18:531 536, 1990.
5. JP van Dijken, M Veenhuis. Eur J Appl Microbiol, 27. E Varela, MJ Martinet, AT Martinez. Biochim
9:275, 1980. Biophys Acta 1481:202 208, 2000.
6. T Yoshimura, T Isemura. J Biochem (Tokyo) 69:839, 28. HJ Hecht et al. J Mol Biol 229:153 172, 1993.
1971. 29. G Wohlfahrt et al. Acta Cryst D55:969 977, 1999.
7. JT Pronk, PR Levering, W Olijve, JP van Dijken. 30. S Witt, G Wohlfahrt, D Schomburg, HJ Hecht, H M
Enzyme Microb Technol, 11:160, 1989. Kalisz. Biochem 347:553 559, 2000.
8. D Keilin, EF Hartree. J Biochem 42:221 229, 1948. 31. P Vaha-Vahe. Food Technol Int Eur, p 139, 1994.
9. D Scott. Enzymes in Food Processing. New York: 32. Z Jiang, B Ooraikul. J Food Prod Preserv 13:175 186,
Academic Press, 1975, p 228. 1989.
10. Q H Gibson, BEP Swoboda, V Massey. J Biol Chem 33. N Low et al. J Food Sci 54:118 121, 1989.
239:3927, 1964. 34. CA Kannt et al. J Food Sci 58:104 107, 1993.
11. K Takegawa, K Fujiwara, S Wahora. Biochem Cell 35. YH Lin. Dissertation, University of Rhode Island,
Biol 67:460, 1989. 1987.
12. HM Kalisz, J Hendle, RD Schmid. Appl Microbiol 36. M Dedek, J Hanus, M Vedlich. Int Dairy Cong
Biotechnol 47:502 507, 1997. Sydney, 1970, p 225.
13. D Scott, J Agric Food Chem, 1:727, 1953. 37. Ough, Mcleod. Am J Enol Viticult 26:30 36, 1975.
14. LA Underkofler. Proc Int Symp on Enzyme 38. U Schobinger, P Durr, R Waldvogel. Fluss Obst
Chemistry, Tokyo and Kyoto 1957. London: 59:586 588, 1992.
Pergamon Press, 1957, p 586. 39. AGJ Rand. J Food Sci 37:698 701, 1972.
15. G Tholey, B Wurtz. Soc Biol 159:2512, 1965. 40. M Beverini, AJ Vroemen. European Patent.
16. Catalog PL Biochemicals 1974. 41. Technical Documentation Zendium. Akzo Dental
17. Catalog Serva Biochemicals 1975. Research now Sara Lee, Veenendaal, Netherlands.
18. T Yoshimura, T Isemura. J Biochem (Tokyo) 69:839, 42. M Sandholm et al. J Vet Med B 35:346 352, 1988.
1971. 43. M Desmazeaud et al. Food Ingredients Europe. Conf
19. S Nakamura, S Fujiki. J Biochem (Tokyo) 63:51, Proc Maarssen, Netherlands, 1989, pp 96 103.
1968. 44. -. Luther, U.S. Patent 2783150, 1957.
20. H Tsuge, O Natsuaki, K Ohashi. J Biochem (Tokyo) 45. S Haarasilta, S Vaisanen, D Scott. U.S. Patent
78:835, 1975. 136003, 1987.
21. S Hayashi, S Nakamura. Biochim Biophys Acta 46. S Haarasilta, T Pullinen, S Vaisanen, I Tammersalo-
438:37 48, 1976. Karsten. U.S. Patent 4990343, 1991.
22. S Hayashi, S Nakamura. Biochim Biophys Acta 47. DSM-Gist, P.O. Box 1, 2600 MA Delft, Netherlands.
657:40 51, 1981. Technical Documentation Hemilox.
23. RL Kelley, CA Reddy. J Bacteriol 166:269, 1986. 48 J Raba, HA Mottola. Crit Rev Anal Chem 25:1 42,
24. M Kriechbaum et al. FEBS Lett 255:63 66, 1989. 1995.
Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved.