Rare keto-aldoses from enzymatic oxidation: Substrates and oxidation products of pyranose 2-oxidase

Article abstract of DOI:10.1002/(SICI)1521-3765(19981204)4:12<2442::AID-CHEM2442>3.0.CO;2-A

Pyranose oxidases are known to oxidise D-glucose, D-xylose and L- sorbose to keto-aldoses, biochemically interesting compounds that may also be used for synthetic purposes in a variety of reactions. In this study pyranose oxidase from the basidiomycete Peniophora gigantea was investigated, and it was found that this enzyme is able to oxidise a broad variety of substrates very effectively. In analogy to its natural mode of action, most substrates are oxidised regioselectively in position 2. Certain compounds, however, are converted into 3-keto derivatives, and the enzyme even exhibits transfer potential, that is, disscharides are formed from β-glycosides of higher alcohols. Substrates that may be oxidised at C-2 in yields between 40-98% are D-allose, D-galactose, 6-deoxy-D-glucose, D-gentiobiose, α-D-glucopyranosyl fluoride and the very interesting 3-deoxy-D-glucose. 1,5-Anhydro-D-glucitol (1-deoxy-D-glucose) is very effectively oxidised in position 2 in 98% yield and additionally gives a product of dioxidation at C-2 and C-3 upon prolonged reaction time Selective oxidation at C-3 was found for 2-deoxy-D-glucose in very good yields and for methyl β-D-gluco- and methyl β-galactopyranoside in lower yields. All oxidation products were unequivocally characterised by NMR spectroscopy and/or chemical derivatisation. In addition, the kinetic data of the enzymatic reactions were determined for all substrates. On the basis of these data and the structural characteristics of the substrates, a model for the minimal structural requirements of the enzyme-substrate interaction is suggested. The enzyme presumably uses two different binding modes for the regioselective C-2 and the C-3 oxidations, which are described.

Full text of DOI:10.1002/(SICI)1521-3765(19981204)4:12<2442::AID-CHEM2442>3.0.CO;2-A

FULL PAPER
Rare Keto-Aldoses from Enzymatic Oxidation: Substrates and Oxidation
Products of Pyranose 2-Oxidase**
Stefan Freimund, Alexander Huwig, Friedrich Giffhorn, and Sabine Köpper*
Abstract: Pyranose oxidases are known
to oxidise d-glucose, d-xylose and l-
sorbose to keto-aldoses, biochemically
interesting compounds that may also be
used for synthetic purposes in a variety
of reactions. In this study pyranose
oxidase from the basidiomycete Penio-
phora gigantea was investigated, and it
was found that this enzyme is able to
oxidise a broad variety of substrates very
effectively. In analogy to its natural
mode of action, most substrates are
oxidised regioselectively in position 2.
Certain compounds, however, are con-
verted into 3-keto derivatives, and the
enzyme even exhibits transfer potential,
that is, disaccharides are formed from b-
glycosides of higher alcohols. Substrates
that may be oxidised at C-2 in yields
between 40 ± 98% are d-allose, d-galac-
tose, 6-deoxy-d-glucose, d-gentiobiose,
a-d-glucopyranosyl fluoride and the
very interesting 3-deoxy-d-glucose. 1,5-
Anhydro-d-glucitol (1-deoxy-d-glucose)
is very effectively oxidised in position 2
in 98% yield and additionally gives a
product of dioxidation at C-2 and C-3
upon prolonged reaction time. Selective
oxidation at C-3 was found for 2-deoxy-
d-glucose in very good yields and for
methyl b-d-gluco- and methyl b-galac-
topyranoside in lower yields. All oxida-
tion products were unequivocally char-
acterised by NMR spectroscopy and/or
chemical derivatisation. In addition, the
kinetic data of the enzymatic reactions
were determined for all substrates. On
the basis of these data and the structural
characteristics of the substrates, a model
for the minimal structural requirements
of the enzyme ± substrate interaction is
suggested. The enzyme presumably uses
two different binding modes for the
regioselective C-2 and the C-3 oxida-
tions, which are described.
Keywords: carbohydrates ´ enzyme
catalysis ´ glycosides ´ oxidases ´
oxidations
other substrates such as d-galactose, d-allose[7, 8] and 6-deoxy-
d-glucose[8] have also been reported. However, none of these
reports delivered quantitative data with respect to the identity
of the products: either only the activity of the enzyme was
demonstrated through the liberation of H2O2 or the products
were characterised only indirectly.
Introduction
Pyranose oxidases are widely distributed among white-rot
fungi,[1, 2] in which they contribute to the ligninolytic system by
producing H2O2.[3] Pyranose oxidases catalyse the regioselec-
tive oxidation of d-glucose at carbon 2, giving the 2-keto-
aldose d-glucosone (2) (Scheme 1). Up to now six pyranose
oxidases have been purified to homogenity from fungal
sources.[1, 4, 5] The existence of ample patent literature on
pyranose oxidases reflects the importance of these enzymes
for biotechnological applications, of which the transformation
of d-glucose to d-fructose is the most prominent one.[6]
Pyranose oxidases from different sources are also known to
oxidise l-sorbose and d-xylose, and the conversions of some
The reaction products of oxidations with pyranose oxidases
are dicarbonyl sugars, a class of biologically as well as
chemically important compounds. Dicarbonyl sugars are, for
example, found in mammalian tissues, in which they are
involved in the cross-linking of proteins and thus affect tissue
functions. d-Glucosone (2) is assumed to be formed by
autooxidation of glucose[9] or other processes in mammals,[10]
and the biochemically most important 3-deoxy-d-glucosone
(18) occurs in blood and urine[11] in concentrations that are
elevated in diabetes.[12] 1,5-Anhydro-d-fructose (20) is be-
lieved to be a metabolite of 1,5-anhydro-d-glucitol (12), which
is present in cerebrospinal fluids and plasma, and plays a role
in diabetes.[13] With regard to their chemical use, dicarbonyl
sugars represent interesting synthons because they combine
the high number of functional groups and the inherent chiral
information of sugars with a higher degree of chemical
diversity in functional groups than normal. They have the
potential to be used in a variety of chemical reactions and thus
[*] Dr. S. Köpper, Dr. S. Freimund
Fachbereich Chemie, Postfach 2503, D-26121 Oldenburg (Germany)
Fax : (‡ 49)441-798-3329
E-mail: koepper@chemie.uni-oldenburg.de
Prof. Dr. F. Giffhorn, Dr. A. Huwig
Angewandte Mikrobiologie, Universität des Saarlandes
Postfach 151150, D-66041 Saarbruecken (Germany)
[**] Presented in part at the XVIII International Carbohydrate Sympo-
sium, Milano (Italy), July 21 ± 26, 1996, poster BP085.
2442
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0412-2442 $ 17.50+.50/0
Chem. Eur. J. 1998, 4, No. 12

2442±2455
Scheme 1. Oxidation of some suitable substrates by pyranose oxidase (PROD).
give valuable products.[14, 15] In particular, certain dicarbonyl
sugars were used for the efficient synthesis of antibiotics,[8, 16]
pyrrolidine and piperidine aminosugars.[17] With respect to
Abstract in German: Pyranose-Oxidasen oxidieren d-Gluco-
se, d-Xylose, und l-Sorbose zu biochemisch interessanten
Keto-aldosen, die sich auûerdem für synthetische Anwendun-
gen in verschiedensten Reaktionen eignen. In diesem Beitrag
wurde die Pyranose-Oxidase aus dem Basidiomyceten Penio-
phora gigantea untersucht und es wurde festgestellt, daû dieses
Enzym eine breite Vielfalt unterschiedlicher Substrate sehr
effektiv oxidiert. ¾hnlich der natürlichen Aktivität werden die
meisten Substrate dabei an Position 2 oxidiert. Einige der
Verbindungen lassen sich jedoch in 3-Keto-Derivate über-
führen und das Enzym zeigt auûerdem Transfer-Potential: z.B.
werden Disaccharide aus b-Glycosiden höherer Alkohole
gebildet. Unter den Substraten, die in 40-98% Ausbeuten an
C-2 oxidiert werden, finden sich d-Allose, d-Galactose,
6-Desoxy-d-Glucose, d-Gentiobiose, a-d-Glucopyranosyl-
fluorid und die interessante 3-Desoxy-d-Glucose. 1,5-Anhy-
dro-d-Glucitol (1-Desoxy-d-Glucose) wird in 98% an C-2
oxidiert und läût sich bei längeren Reaktionszeiten zusätzlich
in ein an C-2 und C-3 zweifach oxidiertes Produkt überführen.
Eine selektive Oxidation ausschlieûlich an C-3 erfolgt an
2-Desoxy-d-Glucose in sehr guten Ausbeuten, während Me-
thyl-b-d-Gluco- und Methyl b-d-Galactopyranosid geringere
Ausbeuten liefern. Alle Produkte wurden eindeutig mit Hilfe
von NMR-Spektroskopie und/oder chemischer Derivatisierung
charakterisiert und die enzymkinetischen Daten wurden für
alle Substrate bestimmt. Auf der Basis dieser Daten und der
strukturellen Eigenschaften der Substrate wird ein Modell der
für eine Enzym/Substrat-Wechselwirkung minimal notwendi-
gen Gruppierungen vorgeschlagen. Das Enzym verwendet
vermutlich zwei verschiedene Arten der Substratbindung für
die regioselektive Oxidation an C-2 und C-3, die beschrieben
werden.
these applications, we were interested in producing
a
greater variety of these compounds by enzymatic con-
versions.
For this purpose, pyranose oxidase from Peniophora
gigantea was used as a biocatalyst. This enzyme oxidizes
d-glucose, d-xylose, l-sorbose and some other hexoses;[5] it
exhibits high stability upon immobilisation to Eupergit C, and
it was used in efficient syntheses of keto-aldoses from
d-glucose, d-xylose and l-sorbose on a preparative scale.[5c, 18]
.
Also, an efficient chemo-enzymatic synthesis of d-tagatose by
enzymatic oxidation of d-galactose with pyranose oxidase was
devised.[15] In this study we elucidate the catalytic potential of
pyranose oxidase from Peniophora gigantea by extensive
chemical characterisation of the reaction products with NMR
spectroscopy and/or chemical derivatisation. A broad range of
carbohydrate substrates are oxidised very effectively by the
enzyme, revealing different regioselectivities in the oxidation
process. The structures of the substrates and the kinetic data
allow the elucidation of the minimal structural requirements
of pyranose oxidase as well as a description of the different
binding modes that are responsible for the observed regiose-
lectivities of the enzymatic oxidation.
Results and Discussion
Substrate studies: As already suggested by Janssen and
Ruelius,[1] the essential structural requirements of substrates
for pyranose oxidase are the six-membered ring of pyranoid
monosaccharides and an equatorially oriented 2-OH group as
Chem. Eur. J. 1998, 4, No. 12
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0412-2443 $ 17.50+.50/0
2443

S. Köpper et al.
FULL PAPER
in d-glucose (1). Accordingly, the natural substrates of
pyranose oxidase, d-glucose (1) and d-xylose (3), which are
abundant in wood, are readily oxidised to the corresponding
2-keto derivatives (Scheme 1). Analogously, l-sorbose (5) is
oxidised at position 5 to give the 5-keto derivative (6), because
this keto-hexose can be considered to contain the same
stereochemical environment as d-glucose with respect to the
orientation of all but the anomeric hydroxyl groups (see
Scheme 1). In order to investigate the actual catalytic
potential of pyranose oxidase and the structural requirements
for substrate conversion, a variety of carbohydrates were
tested for activity with pyranose oxidase from Peniophora
gigantea.
and 1 as in 1-deoxy-glucose (1,5-anhydro-d-glucitol, 12) and
6-deoxy-d-glucose (11) result in a decrease of the catalytic
efficiencies of about one order of magnitude (Table 1), but
quantitative conversions of both substrates are still observed.
1,5-Anhydro-d-glucitol may be considered a derivative of l-
sorbose (5) (Scheme 1), and is converted into the 1,5-anhydro-
d-fructose (20) with a similar catalytic efficiency as l-sorbose.
Probing the specificity of the enzyme for the anomeric
position, it was found that a-d-glucosyl fluoride (13) was
converted with a catalytic efficiency in the range of 1,5-
anhydro-d-glucitol (Table 1). Unfortunately, the b-derivative
of 13 was hydrolysed during the enzymatic reaction, and we
were therefore not able to obtain any data on the enzymatic
acceptance of this compound. In addition, the substrates a-d-
glucose-1-phosphate, d-glucosamine and 3-O-methyl-d-glu-
cose were not accepted by the enzyme.
Position 6 of d-glucose does not seem to be important for
the recognition process. In addition to a deoxygenation, this
position may even contain a sterically demanding sugar unit.
Thus, the disaccharide d-gentiobiose (14) is oxidised easily
with a 80% conversion, but has less favourable kinetic data.
Maltose and cellobiose were not oxidised in their disacchari-
dic form, but we isolated d-glucosone (2) when these
disaccharides were treated with the enzyme. Whether hydrol-
ysis of the glycosidic bond had taken place before or after
oxidation could not be determined unequivocally. During an
NMR spectroscopic study of the reaction we only detected the
intact disaccharide and d-glucosone. Disaccharides d-treha-
lose (Glc-a1,a1-Glc), d-palatinose (Glc-a1,6-Fru) d-meli-
biose (Gal-a1,6-Glc) and d-isomaltose (Glc-a1,6-Glc) did
not react with the enzyme.
Oxidation at C-2: In accordance with a previous study[1] it was
found that pyranose oxidase does not oxidise axial hydroxyl
functions at position 2. This is evidenced by the fact that d-
mannose, d-altrose, d-arabinose and d-lyxose and others that
will follow are not oxidised at all. On the other hand, hexoses
like d-allose (7) and d-galactose (8) with axial OH groups in
positions 3 and 4 are readily oxidised at C-2, although these
substrates were converted with decreased catalytic efficien-
cies (kcat/km) relative to D-glucose (Table 1). Nevertheless, the
chemical conversion of these hexoses into the 2-keto deriv-
Table 1. Kinetic data of enzymatic oxidations with immobilised pyranose
2-oxidase. d-Man, d-alt, d-tal, d-ara, d-lyx, methyl a-d-glc, methyl a-d-gal,
methyl a-d-man, methyl b-d-all, methyl b-d-man, 3-O-methyl-d-glc, a-d-glc-1P,
2-amino-2d-d-glc, 1d-d-nojirimycin, d-palatinose, d-melibiose, d-isomaltose,
d-trehalose were inactive.
[b]
Substrate
kM
[mm]
vmax
[Ug
kcat
[s 1 g
kcat/kM
[s 1 mol
Rel. act.
] [%]
1
1
1
]
]
Oxidation at C-3: All of the substrates so far mentioned were
regioselectively oxidised at position C-2 by pyranose oxidase.
With the 1,5-anhydro d-glucitol (12), however, an additional
product was observed upon prolonged incubation and in the
presence of increased amounts of enzyme. The new product
was identified as the 2,3-diketo derivative 21[19] which resulted
from subsequent oxidation of the 2-keto derivative 20.[20]
During oxidation of the corresponding 4-epimer, 1,5-anhy-
dro-d-galactitol, the same kind of initial 2-oxidation with
subsequent 2,3-dioxidation was observed in a qualitative
experiment.[21]
Oxidation at position 3 was shown to be the only reaction
observed with substrates 24, 25 and 26. Although 2-deoxy-d-
glucose (24) does not possess an oxidisable hydroxyl group in
position 2, it is oxidised with considerable efficiency (Table 1)
to the 3-keto compound 27.[7c] The same regioselective
3-oxidation was found for methyl b-d-glucose (25) and methyl
b-galactose (26), however, with less favourable kinetics and
low yields of the 3-oxidised products 30[22] and 31[23] (Table 2).
1
3
7
8
d-Glc[a]
1.0
28.6
112.3
11.3
[c]
1.0
6.0
9.5
20.0
10.2
8.0
1.6
[c]
19.2
18.3
15.0
106.7
54.6
42.7
8.3
106700
1909
380
735
[c]
102200
16343
8401
100
51
40
8
[c]
96
92
75
d-Xyl[a]
d-All
d-Gal[a]
9
d-Rib
[c]
10
11
12
3d-d-Glc
6d-d-Glc
1,5-Anhydro-
d-glucitol
a-d-GlcF
d-Glcb(1-6)-
d-Glc
102.2
98.1
79.8
13
14
7.9
65.7
13.4
10.3
71.6
54.8
9061
835
67
51
24
25
26
2d-d-Glc
51.5
288.8
33.1
10.3
8.4
0.2
55.0
45.0
1.0
1067
156
52
42
1
Methyl b-d-Glc[a]
Methyl b-d-Gal
0.03
[a] Ref. [5b, 5c], [19]. [b] Referred to purified enzyme bound to 1 g Eupergit C
(d.w.).[18] [c] Reaction too slow for determination.
atives 15 and 16 is achieved in 94 and 70% yield, respectively
(Table 2). On the other hand, the 3-OH axially oriented d-
ribose (9) is oxidised only to a very low extent (5%) in a very
slow reaction, which did not allow the kinetic data of this
reaction to be determined.
Deoxygenation in position 3 as in 3-deoxy-d-glucose (10)
does not significantly alter the catalytic efficiency for this
substrate relative to d-glucose (Table 1). Consequently,
quantitative conversion was observed with this substrate
(Table 2). On the other hand, deoxygenations in positions 6
Glycosyl transfer: Other than methyl glycosides, b-glycosides
of higher alcohols like hexyl-, phenyl-, o- and p-nitrophenyl
are not oxidised by the enzyme, but disaccharides are formed
when these glycosides are treated with less enzyme than usual.
We thus isolated (1 !6)-linked disaccharides 36,[24] 37,[25] 38
and 40[26] in yields between 7 and 20% in addition to minor
amounts of (1 !3)-linked disaccharides 39 and 41 which
2444
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0412-2444 $ 17.50+.50/0
Chem. Eur. J. 1998, 4, No. 12

Pyranose 2-Oxidase
2442±2455
Table 2. Reaction products and conversion rates.
d-Allosone (15) was previously obtained
by enzymatic oxidation[7c] but no data were
available for an unequivocal structural assign-
ment. We obtained d-allosone in a purity of
94% from the enzymatic reaction and were
able to purify the compound by repeated
chromatographic separation. Isopropylidena-
tion of 15 yielded a mixture of the bicyclic
monoisopropylidene 42 and a triisopropyli-
dene derivative 44 in 38% and 4% yield,
respectively (see Scheme 2). Evidence for the
bicyclic double acetal structure of 42 and 43
was given by the chemical shifts for the C-2
carbons at 101 ± 103 ppm in addition to the
downfield shifts for C-5, as well as the very
profound sensitivity of the C-6 resonances for
the anomeric configuration. The 1H NMR
coupling data and chemical shifts are in
agreement with bicyclic structures that
have previously been described for keto-
Substrate
Conver- Product
sion [%]
Oxidation at C-2
7
8
d-All
d-Gal
94
70
d-ribo-hexos-2-ulose
d-lyxo-hexos-2-ulose
15
16
17
18
19
20
21
9
d-Rib
5
100
100
d-erythro-pentos-2-ulose
3d-d-erythro-hexos-2-ulose
6d-d-ribo-hexos-2-ulose
10
11
12
3d-d-Glc
6d-d-Glc
1,5-Anhydro-d-glucitol[a] 100
1,5-anhydro-d-fructose
33
1,5-anhydro-d-erythro-hex-2,3-diulose[b]
13
14
a-d-GlcF
d-Glc b(1-6)- d-Glc
40
80
a-d-arabino-hexos-2-ulo-pyranosyl fluoride 22
d-Glc b(1-6)-d-arabino-hexos-2-ulose
23
Oxidation at C-3
24
2d-d-Glc[a]
80
65
100
14
2d-d-erythro-hexos-3-ulose or:
1d-d-ribulose[c]
27
28[c]
29[c]
30
2d-d-arabino-hexonic acid[c]
methyl b-d-ribo-hexos-3-ulopyranoside
methyl b-d-xylo-hexos-3-ulopyranoside
25
26
Methyl b-d-glc
Methyl b-d-gal
18
31
Transfer reaction
b-Monosaccharide
Disaccharide
32
33
34
Hexyl b-d-Glc
Phenyl b-d-Glc
2-Nitrophenyl b-d-Glc
9
7
15
hexyl b-d-Glc b(1-6)-d-Glc
phenyl b-d-Glc b(1-6)-d-Glc
2-nitrophenyl b-d-Glc b(1-6)-d-Glc and
2-nitrophenyl b-d-Glc b(1-3) d-Glc (4:1)
4-nitrophenyl b-d-Glc b(1-6) d-Glc and
4-nitrophenyl b-d-Glc b(1-3) d-Glc (2:1)
36
37
38
39
40
41
aldoses,[28, 29]
2-keto-glycopyranosides,[30]
aqueous solutions of d-allosone (15) and
3-deoxy-glucosone (18) and in organic solu-
tions of d-glucosone (2).[27] Hydrogenation of
d-allosone (15) with Pd/C afforded a mixture
of d-psicose,[27] d-allose and d-altrose[31, 32] in
35
4-Nitrophenyl b-d-Glc
24
[a] Different products may be formed when reaction conditions are changed. [b] Side product
formed upon prolonged reaction. [c] Products formed when less than 20000 U catalase are used
in the enzymatic reaction.
1
a ratio of 5:4:1 according to H NMR spec-
troscopy (Table 3).
clearly demonstrate the transfer potential of the enzyme
(Table 2). In these reactions d-glucosone was formed in a side
reaction.
Treatment with the normal amounts of enzyme led to
hydrolysis of the glycosidic linkages which had already been
observed in the case of the b-linked disaccharides maltose and
cellobiose. Since we were only interested in elucidating the
catalytic potential of pyranose oxidase, we have not yet
optimised the transfer reactions, but higher amounts of
(1 !6)- and (1 !3)-linked disaccharides may presumably be
formed when better reaction conditions are applied. Both
(1 !6)- and (1 !3)-linked disaccharides were isolated and
characterised following per-O-acetylation.
Scheme 2. Isopropylidenation of 15.
Synthesis of d-galactosone (16) has been reported by
several groups both enzymatically[7b, 7c] and chemically.[33] The
compound consists of at least 15 different isomers in water,
which will be reported elsewhere.[27] The position of oxidation
in 16 has previously been identified by us through hydro-
Chemical characterisation of keto-aldoses and subsequent
products: The products were isolated and characterised
chemically either directly by NMR spectroscopy and/or
chemical derivatisation. Keto-aldoses tend to adopt compli-
cated isomeric equilibria of different forms. Consequently the
NMR spectroscopic assignment is quite complicated and
chemical reactions lead to complex mixtures. Some of our
derivatives even form as many as 15 different isomers in
aqueous solution, and we preferred to have an independent
characterisation of the position of oxidation by chemical
derivatisation and/or hydrogenation of the keto group. A full
description of the composition of keto-aldoses featuring all
NMR spectroscopic data will be given in a different paper.[27]
Here the derivatives are represented in their pyranoid,
hydrated forms, which represent the most abundant isomeric
form in water.
genation of 16 to d-tagatose,[15] whichÐamong others[34]
Ð
proved to be an effective way for the production of this rare
sugar. Upon treatment with acetone/FeCl3 a number of
derivatives were formed and the triisopropylidene derivative
45 was isolated and characterised. Enzymatic oxidation of d-
ribose showed a conversion of only 5%. The oxidation
product d-ribosone (17) was therefore not isolated, but could
be identified by NMR spectroscopy in comparison to
literature values.[35]
The 6-deoxy derivative of glucose was converted quantita-
tively into 6-deoxy-d-glucosone (19), a compound that has
previously been obtained enzymatically and described as its
diphenylhydrazone.[8] We characterised 19 first directly by
NMR spectroscopy, followed by an isopropylidenation that
Chem. Eur. J. 1998, 4, No. 12
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0412-2445 $ 17.50+.50/0
2445

S. Köpper et al.
FULL PAPER
Table 3. Products formed upon catalytic hydrogenation of hexos-2-uloses.
Keto-aldose
Hydrogenation products
Aldoses
Ketose
[%]
[%]
ratio[a]
OH-2 eq/ax
2
d-arabino-hexos-2-ulose
d-threo-pentos-2-ulose
d-ribo-hexos-2-ulose
d-fru
85
80
50
76
85
1
d-glc
d-man
d-xyl
d-lyx
d-all
d-alt
d-gal
d-tal
3d-d-glc
3d-d-man
6d-d-man
glcb(1-6)glc
glcb(1-6)man
15
20
50
24
15
1:4
1:1
4:1
1:3
1:1
4
99
d-xylulose
d-psi
3
15
16
18
101
102
103
7
d-lyxo-hexos-2-ulose
d-tag
8
3d-d-erythro-hexos-2-ulose
3d-d-fru
10
19
23
6d-d-arabino-hexos-2-ulose
Glcb(1-6)-d-arabino-hexos-2-ulose
6d-d-fru
glc b(1-6)-d-fru
50
33[b]
50
67[b]
75
14
1:1
[a] Ratio of hydrogenation products with equatorial and axial OH-2. [b] Ratio determined after acetylation of the hydrogenation products.
yielded the diisopropylidene derivative 46[36] among a variety
of unidentified products (Scheme 3). Hydrogenation of 19
afforded a mixture of 6-deoxy-d-mannose and 6-deoxy-d-
fructose in the ratio of 1:1 (according to NMR, see Table 3).
es.[40, 41, 43] From the oxidation with pyranose oxidase we
isolated 1,5-anhydro-d-fructose (20) in a yield of 98%; the
product is thus available in three steps from d-glucose via
2,3,4,6-tetra-O-acetyl-a-d-glucopyranosyl bromide, reduction
at position 1,[44] and enzymatic oxidation in an overall yield of
approximately 60%. When the enzymatic oxidation was
prolonged a product of dioxidation formed in 33% yield.
This new diketo compound 21 was unequivocally character-
ised by NMR; nevertheless, it was not possible to purify 21
from this mixture and we therefore prepared O-methyl
oximes (Scheme 4), which led to the pure dioxime 48 (four
stereoisomers in a ratio of 66:30:3:1) after separation from
monoxime 47. The latter could also be prepared from pure 20
in a yield of 90% (only one stereoisomer).
Scheme 3. Isopropylidenation of substrates 16 ± 19.
3-Deoxy-d-glucose (10) is quantitatively oxidised to its
2-keto compound 18. In mammals 18 is presumably formed by
elimination reactions from glycated Maillard products and is
therefore present in urine and blood.[11, 12] Consequently,
several synthetic pathways leading to this sugar have been
developed giving overall yields of 40% from d-glucose.[37]
Here by means of isopropylidenation of d-glucose, reduc-
tion,[38] deprotection, and enzymatic oxidation, the pure
compound 18 is available in an overall yield of 70%.
Hydrogenation of 18 with Pd/C mainly gives the product of
C-1 reduction, 3-deoxy-d-fructose (85%), an equally attrac-
tive synthon for biochemical investigations, in combination
with 3-deoxy-d-mannose and 3-deoxy-d-glucose (15%, 1:1
Scheme 4. Synthesis of O-methyl oximes from 1,5-anhydro-d-erythro-hex-
2,3-diulose (21).
1
according to H NMR[39] of the raw product, see Table 3).
Enzymatic oxidation of 1,5-anhydro-d-glucitol (12) to 1,5-
anhydro-d-fructose (20) was quantitative. We have already
shown that 20 forms dimers in nonaqueous solutions[40] and
may be used for elimination reactions.[40, 41] Compound 20 is a
well-known product, which was first synthesised by Lich-
tenthaler et al.[14] and later prepared enzymatically.[13, 42] The
synthetic interest stems from the fact that 20 presumably
represents the biosynthetic precursor of the antibiotic Micro-
thecin,[11b] and that it may be used for synthetic purpos-
a-d-glucosyl fluoride (10) was converted enzymatically into
2-keto a-d-glucosyl fluoride (22), but the raw product of the
reaction also contained d-glucosone in 30% yield. Purifica-
tion of the fluorohexosulose was difficult because 22 hydro-
lysed on the column. The product was obtained with severe
losses in a purity of only 90% and was characterised
spectroscopically.
The disaccharide d-Glcb(1 !6)d-Glc (d-gentiobiose) (14)
was oxidised in the reducing ring by pyranose oxidase to yield
2446
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0412-2446 $ 17.50+.50/0
Chem. Eur. J. 1998, 4, No. 12

Pyranose 2-Oxidase
2442±2455
the disaccharide 23. Purification again proved to be difficult;
we therefore characterised the product by NMR spectroscopy
and found that 23 in aqueous solution forms the two pyranoid
anomers 23a and 23b (85%) and a mixed pyranoid ± furanoid
form 23c in 15% yield. Further characterisation of the
product was accomplished by hydrogenation and subsequent
acetylation to a mixture of nearly equal amounts of the per-O-
acetylated derivatives of b-d-glucopyranosyl-(1-6)a/b-d-fruc-
tofuranose (51), d-gentiobiose (49)[45] and b-d-glucopyrano-
syl-(1-6)a/b-d-mannose (50)[46] (Scheme 5).
the other hand, exclusively adopts the usual hydrated form in
water.[30]
Enzymatic oxidation of 2-deoxy-d-glucose (24) gives the
3-keto derivative 27 (Scheme 7). Although 27 has been
reported as a probable enzymatic oxidation product of 24,[7c]
it has never been characterised or synthesised by chemical
Scheme 7. Enzymatic oxidation of 2-deoxy-d-glucose (24) to give the
3-keto derivative 27.
means to our knowledge. Also, when we attempted a chemical
synthesis of 27 from the glycoside 52[47] this proved to be
difficult. The product 27 easily decomposed or was oxidised
by molecular oxygen and subsequently decarboxylated to
form the pentulose 28[48] under the reaction conditions. In
addition we observed that the 2-deoxy-3-keto-hexose (27)
exchanges protons H-2 in deuterated solvents, a behaviour
that was also reported for 3-keto-d-glucose[49] and therefore
should create the potential for attractive chemistry with this
compound.
Although the enzymatic oxidation of 2-deoxy-d-glucose
(24) to the 3-keto compound 27 is a straightforward process
with an isolated yield of 75%, we initially obtained confusing
results when the enzymatic oxidation was carried out under
standard conditions. We isolated mixtures of the 3-keto
derivative 27 with the lactones 53 and 54[50] or the acid
29,[50, 51] and even the pentulose 28[48] in differing ratios
(Scheme 7). 1-Deoxy-d-ribulose 28 was characterised by
NMR spectroscopy, and the C-1 oxidation products were
identified by acetylation to the derivatives 55,[52] 56[53] and
57.[54]
Scheme 5. Characterisation of 23 was accomplished by hydrogenation and
subsequent acetylation to a mixture of nearly equal amounts of the per-O-
acetylated derivatives of b-d-glucopyranosyl-(1-6)a/b-d-fructofuranose
(51), d-gentiobiose (49)[45] and b-d-glucopyranosyl-(1-6)a/b-d-mannose
(50).
Chemical identification of 30[22] and 31,[23] the products of
oxidation in position 3 of b-methyl glycosides, was straightfor-
ward. The gluco derivative 30 exists in a nonhydrated form
30a as well as in a hydrated form 30b in aqueous solution
(30a/30b ˆ 2:1), see Scheme 6. This has already been descri-
bed by Tsuda et al.,[30] but we found that the published
13C NMR data were erroneous. The galacto derivative 31, on
Formation of the pure C-3 oxidation product 27, however,
was achieved when we used two to four times of the usual
amount of catalase to destroy the H2O2 formed during the
oxidation. This demonstrates that the pentulose 28 as well as
the C-1 oxidation products 29, 53 and 54 are formed by
chemical oxidation with H2O2, which is produced during the
enzymatic reaction (or with O2 from the bubbling air ). We
assume that the C-1 oxidated derivatives 29, 53, and 54 are
formed by direct chemical oxidation of 2-deoxy-d-glucose
Scheme 6. Equilibrium between the nonhydrated form (30a) and the
hydrated form (30b) of methyl b-d-ribo-hexose-3-ulopyranoside (30).
Chem. Eur. J. 1998, 4, No. 12
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0412-2447 $ 17.50+.50/0
2447

S. Köpper et al.
FULL PAPER
(24), by H2O2 (or with O2 from the bubbling air), which is in
agreement with the usual chemical oxidation of 2-deoxy-d-
glucose.[50, 51] On the other hand, the product 28 may be
formed by subsequent chemical oxidation of enzymatically
oxidised 27. The latter may proceed by oxidation of 27 to a b-
keto-carboxylic acid A, which should easily decarboxylate to
form the keto/enol structures B and isolated 28 (Scheme 8).
Evidence for this mechanism is given by the fact that the
isolated 1-deoxy-d-ribulose (28) is contaminated with its C-3
epimer 58[55] (6%), and the observation that the pentulose 28
is formed in our attempts to synthesise 27 chemically from 52;
the latter could be influenced by bubbling air through the
reaction mixture.[47]
Since no contaminating enzymes or protein were detected,
the hydrolytic/transfer activity is probably caused by a side
activity of the enzyme. This is additionally evidenced by the
fact that both a- and b-glucosidic linkages were hydrolysed,
which is unlikely to occur by contamination of usually most
selective glucosidases. The hydrolytic activity of PROD was
1
determined to be 0.5 Umg for b-glucosidic linkages and
1
0.02 Umg for a-glucosidic linkages.
The most interesting activity of pyranose 2-oxidase is
obviously the oxidation process, which represents both the
biochemically important function and an attractive synthetic
possibility. Pyranose 2-oxidase is a flavoprotein, with presum-
ably a tryptophan involved in the binding of the substrate,
the flavin adenine dinucleotide (FAD) or the oxygen.[5c]
d-Glucose is the natural and so far optimal substrate for the
enzyme with respect to a high affinity and fast oxidation
process. We found that 3-deoxy-d-glucose (10) is accepted
with the same high affinity and fast oxidation, and the 6-deoxy
derivative 11 shows a comparable relative activity, but slightly
worse affinity (see Table 1). Moreover, positions 3 and 4 of the
substrate may bear axial as well as equatorial hydroxyl groups.
Therefore no essential binding should occur at positions 3, 4
or 6. Also, a deoxygenation in position 1 is accepted and the
fact that freshly prepared a-d-glucose and b-d-glucose show
the same rate constants[56] indicates that there is no essential
binding at position 1. With the finding that deoxynojirimycin
is not active, the ring oxygen becomes essential, which is also
in agreement with previous data for comparable enzymes[1]
and led to the naming of the enzyme as pyranose oxidase. We
therefore propose that passive binding may occur to the ring
oxygen in addition to the dynamic binding to position 2 in the
natural action of the enzyme, the 2-oxidation process (see
Figure 1).
Scheme 8. Proposed mechanism for the formation of 28 from 27. This
reaction could proceed by oxidation of 27 to a b-keto-carboxylic acid A,
which should easily decarboxylate to form the keto/enol structures B and
isolated 28.
Nevertheless, since the affinities to 6-deoxy-d-glucose (11),
d-xylose (3) and the 1-deoxy compound 12 are reduced with
respect to the natural substrate (see Table 1), there may be
additional stabilising interactions to positions 1 and 6, but
these do not seem to be essential. There certainly are some
steric hindrances in the binding site, especially near position 3
of the substrate, which is indicated by the fact that 3-O-
methyl-d-glucose is not accepted and by the extremely high
kM for d-allose.
Conclusions
Pyranose 2-oxidase (PROD) from Peniophora Gigantea
accepts a broad variety of carbohydrate substrates. The kind
of enzymatic action, however, is not identical for all sub-
strates. We observed regioselective oxidations in position 2
and in position 3, as well as hydrolysis of glycosidic linkages in
disaccharides and transfer of glycosyl
When there is no oxidisable function at position 2, the
enzyme dynamically binds position 3 instead; this occurs for
2-deoxy-d-glucose (24) and the oxidised 1,5-anhydro-d-fruc-
units to form disaccharides. This sug-
gests that the enzyme exhibits at least
two different activities: oxidation and
hydrolysis/transfer. Multiple activities
in one protein are observed with nu-
merous enzymes; especially well-docu-
mented examples for this are the en-
zymes involved in polyketide synthe-
sis.[56] The transfer reaction as well as
the hydrolysis were observed with an
enzyme that had been purified to
homogenity and analysed by polyacry-
lamide gel electrophoresis (PAGE).
Figure 1. Two possible binding modes for substrates of pyranose oxidase: regioselective oxidation in
position 2 (left) and regioselective oxidation in position 3 (right).
2448
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0412-2448 $ 17.50+.50/0
Chem. Eur. J. 1998, 4, No. 12

Pyranose 2-Oxidase
2442±2455
General procedure III (GP III). Isopropylidenation: Dry sugar (0.25 ±
2.5 mmol) was stirred vigorously in dry acetone (3 ± 10 mL) until a fine
suspension had formed. Under a nitrogen atmosphere and with stirring
FeCl3 (20 ± 50 mg) was added, leading to a clear yellow solution. After
stirring overnight and concentration to a volume of 2 mL, the solution was
transferred onto a silica gel column, which was first eluted with n-hexane
and then with n-hexane/acetone (2:1).
tose (20). This modified regioselectivity may be explained
with a substrate that is turned around in the same binding site.
This could possibly occur by a binding of enzymatic groups to
the other free electron pair of the ring oxygen, which now
turns the C-3 hydroxyl group to nearly the same acceptable
distance for dynamic interaction. This mode of binding could
also occur with the b-glycosides 25 and 26, where steric
hindrance of the b-glycosidic methoxyl group might induce
the change in the binding mode. Larger glycosidic groups like
hexyl, phenyl, and nitrophenyl are sterically hindered and
binding becomes impossible in either mode.
The enzyme pyranose 2-oxidase may be used for regiose-
lective oxidations of various substrates. The oxidation of
position 2 is straightforward and proceeds in good yields with
most substrates. Certain substrates, however, may be con-
verted into the 3-keto compounds in equally high yields.
Moreover, with the above outline of the minimal structural
requirements for a compound to function as substrate for
pyranose 2-oxidase, it should be possible to devise other
attractive substrates for regioselective oxidations with this
enzyme.
General procedure IV (GP IV). Catalytic hydrogenation: To a solution of
the sugar (0.1 ± 2.0 mmol) in H2O (5 mL) Pd/C (10%) was added in a ratio
of 4:1 (w/w). The mixture was stirred under hydrogen until TLC showed
complete reaction. It was then filtered over celite, washed with hot H2O,
and evaporated.
Hydrogenation of a/b-d-arabino-hexos-2-ulose (2): Hydrogenation of 2
(115 mg, 0.59 mmol) for 2 d as described in GP IV gave a product (95 mg)
consisting of d-fructose, d-glucose (1) and d-mannose in a ratio of 85:3:12
(NMR).
Hydrogenation of a/b-d-threo-pentos-2-ulose (4): Hydrogenation of 4
(90 mg, 0.54 mmol) for 5 d as described in GP IV gave a product (72 mg)
consisting of d-xylulose,[58] d-xylose (3) and d-lyxose[31a, 58] in a ratio of 8:1:1
(NMR).
a/b-d-Ribo-hexos-2-ulose (15): Treatment of d-allose (7, 700 mg, 3.9 mmol)
with PROD (86 units) for 3 d as described in GP I yielded a material
(730 mg) consisting of 15 (94%) and 7 (6%) (NMR). For analytical
purposes crude 15 (300 mg) was purified by HPLC on a Dowex column
giving a pale yellow foam (110 mg, 35% relative). M.p. 101 ± 1098C
(amorphous); [a]2D0 ˆ ‡16.3 (c ˆ 0.94 in H2O); MS m/z (%): 179 (38)
‡
‡
‡
[M ‡H], 161 (100) [M ‡H H2O], 133 (12) [M ‡H H2O CO], 115
‡
(31) [M ‡H 2H2O CO]; NMR of the predominant isomer b-d-ribo-
Experimental Section
1
hexos-2-ulo-1,5-pyranose (hydrated form): H NMR (D2O): d ˆ 3.506 (dd,
J5,6b ˆ 5.9, J6a,6b ˆ 12.1 Hz,1H, H-6b), 3.603 (AB, 1H, H-5), 3.605 (dd ꢀ AB,
J3,4 ˆ 3.2 Hz, 1H, H-4), 3.687 (2dd, J5,6a ˆ 1.3 Hz, 2H, H-3, H-6a), 4.705 (s,
1H, H-1); 13C NMR (D2O): d ˆ 61.58 (C-6), 66.17 (C-4), 73.51 (C-3), 73.84
(C-5), 92.65 (C-2), 92.74 (C-1); C6H10O6 ´ H2O (196.2): calcd C 36.74, H
6.17; found C 36.74, H 5.89.
General methods: The reactions were monitored by TLC on silica gel
60F254 (Merck) and the spots were detected by UV absorption, staining
with sulfuric acid or with dinitrophenylhydrazine. For column chromatog-
raphy silica gel 60 (70 ± 230 mesh, Merck) was used. Preparative TLC was
performed on 20 Â 20 Â 0.2 cm plates coated with silica gel 60F254 (Merck)
and HPLC on LiChrosorb Si60 (Merck) with a column of 250 mm Â
Hydrogenation of a/b-d-ribo-hexos-2-ulose (15): Hydrogenation of crude
15 (39 mg, 0.20 mmol) for 40 h as described in GP IV gave a product
(31 mg) consisting of d-psicose,[59] d-allose (7) and d-altrose[30a, 32] in a ratio
of 5:4:1 (NMR).
1
25 mm  7 mm with a flow rate of 5 mLmin or on a preparative column
(600  26 mm) of DOWEX 50W-X8 (loaded with Ca2‡) at 718C with H2O
as eluent (flow rate of 9 mLmin 1). Melting points are not corrected and
were recorded on a SM-Lux from Leitz. Optical rotations were determined
with a Perkin ± Elmer model 241MC. Mass spectra were obtained on a
MAT212 from Finnigan (70 eV, He gas support) with chemical ionisation.
NMR spectra were recorded on a Bruker AMX500 spectrometer at
500.14 MHz for 1H and 125.76 MHz for 13C at 300 K. 2D spectroscopy
(phase-sensitive DQF-COSY and HMQC) was performed with the stand-
ard Bruker software. Chemical shifts are referenced to internal acetone
(d ˆ 2.030 and 30.50) for D2O as solvent, in CDCl3 to d ˆ 7.270 and 77.00 or
C6D6 to d ˆ 7.150 and 128.00 ppm for these solvents, respectively.
a/b-d-Erythro-Pentos-2-ulose (17): Treatment of d-ribose (9, 200 mg,
1.33 mmol) with PROD (18 units) for 20 d as described in GP I gave a
material consisting of 9 (95%) and 17 (5%).[35]
3-Deoxy-a/b-d-erythro-hexos-2-ulose (18): Treatment of 3-deoxy-d-ribo-
hexose (10, 100 mg, 0.61 mmol) with PROD (46 units) for 19 h as described
in GP I gave a colourless syrup (108 mg, 98%) consisting of pure 18.[59]
[a]2D0
ˆ
3.4 (after 30 min, c ˆ 4.25 in H2O; ref. [37b]: [a]D25
ˆ
2.5, after
‡
10 min, c ˆ 5.98 in H2O); MS m/z (%): 163 (100) [M ‡H], 145 (22)
‡
[M ‡H H2O]; C6H10O5 ´ H2O (180.1): calcd C 40.00, H 6.71; found C
Pyranose 2-oxidase (PROD, EC 1.1.3.10) was isolated and immobilised as
previously described.[5] The hydrolytic side activity of PROD was
40.62, H 6.64.
1
1
Hydrogenation of 3-deoxy-a/b-d-erythro-hexos-2-ulose (18): Hydrogena-
tion of 18 (108 mg, 0.60 mmol) for 4 d as described in GP IV gave a product
(89 mg) consisting of 3-deoxy-d-erythro-hex-2-ulose, 10[30] and 3-deoxy-d-
arabino-hexose[39] in a ratio of 12:1:1 (NMR).
determined to be 0.5 Umg as a b-glucosidase activity and 0.02 Umg
as
a a-glucosidase activity by means of 2-hydroxymethylphenyl-b-d-
glucopyranoside and 4-nitrophenyl-a-d-glucopyranoside for the assay.[57a]
Catalase (EC 1.11.16) was a generous gift from Novo Nordisk, Bagsvñrd
(Denmark). The kinetic data were determined as described.[5c]
6-Deoxy-a/b-d-arabino-hexos-2-ulose (19): Treatment of 6-deoxy-d-glu-
cose (11, 500 mg, 3.05 mmol) with PROD (25 units) for 28 h as described in
GP I gave a pale yellow foam (450 mg, 81%) consisting of pure 19[32b, 8]
(NMR). M.p. 101 ± 1158C (amorphous); [a]2D0 ˆ ‡16.3 (c ˆ 0.98 in H2O);
General procedure I (GP I). Enzymatic oxidation: Bioconversions with
pyranose 2-oxidase (PROD) were performed in a 500 mL five-neck flask
with gentle stirring and with aeration (0.01 Lmin 1) through a porous
sintered glass tube at 228C. In a final volume of H2O (200 mL) the substrate
was treated with differing amounts of PROD and catalase (10000 U,
0.2 mg) while the pH was maintained at 7.0 by titration with NaOH (0.01 n).
When the conversion was complete, the suspension was filtered through a
sintered glass funnel with a porosity of 16 ± 40 mm to remove immobilised
PROD. Soluble catalase was removed by ultrafiltration through a YM10
membrane (Amicon) and the filtrate was first evaporated at max 358C and
then lyophilized.
‡
‡
MS m/z (%): 163 (10) [M ‡H], 145 (100) [M ‡H H2O], 117 (56)
‡
‡
[M ‡H H2O CO], 99 (18) [M ‡H 2H2O CO]; NMR of the pre-
dominant isomer 6-deoxy-a-d-arabino-hexos-2-ulo-1,5-pyranose (hydrated
form): 1H NMR (D2O): d ˆ 1.067 (d, J5,Me ˆ 6.3 Hz, 3H, Me), 3.089 (dd,
J3,4 ˆ 9.7 Hz, J4,5 ˆ 9.6 Hz, 1H, H-4), 3.502 (d, 1H, H-3), 4.684 (s, 1H, H-1);
13C NMR (D2O): d ˆ 17.01 (Me), 68.10 (C-5), 73.51 (C-3), 74.29 (C-4), 93.93
(C-2), 94.79 (C-1); C6H10O5 ´ H2O (180.1): calcd C 40.00, H 6.71; found C
39.69, H 5.71.
General procedure II (GP II). Acetylation: At 58C dry sugar (0.03 ±
0.6 mmol) in dry pyridine (2 ± 5 mL) was treated overnight with acetic
anhydride (0.5 ± 1 mL). After addition of H2O (5 ± 10 mL), the solution was
extracted with CH2Cl2 and the organic phase was dried and coevaporated
several times with toluene.
Hydrogenation of 6-deoxy-a/b-d-arabino-hexos-2-ulose (19): Hydrogena-
tion of 19 (50 mg, 0.28 mmol) for 5 d as described in GP IV gave a
product (44 mg) consisting of 6-deoxy-d-arabino-hex-2-ulose (d-rhamnu-
lose)[61] and 6-deoxy-d-mannose (d-rhamnose),[30, 57c, 61] in a ratio of 1:1
(NMR).
Chem. Eur. J. 1998, 4, No. 12
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0412-2449 $ 17.50+.50/0
2449

S. Köpper et al.
FULL PAPER
Enzymatic oxidation of 1,5-anhydro-d-glucitol (12):
3.569 (d, 1H, H-3), 3.657 (ddd, J5,6a ˆ 2.2 Hz, 1H, H-5), 3.699 (dd, 1H,
H-6a), 5.205 (d, J1,F ˆ 51.2 Hz, 1H, H-1); 13C NMR (D2O, hydrated form):
a) 1,5-Anhydro-d-fructose (20): Treatment of 1,5-anhydro-d-glucitol (12,
200 mg, 1.22 mmol) with PROD (37 units) for 9 h as described in GP I gave
a colourless foam (204 mg, 98%) consisting of pure 20.[40b] M.p. 107 ± 1128C
d ˆ 60.65 (C-6), 67.94 (C-4), 73.43, 74.78 (C-3 and C-5), 92.47 (d, J2,F
31 Hz, C-2), 108.00 (d, J1,F ˆ 226 Hz, C-1).
ˆ
(amorphous); [a]D20
ˆ
32.9 (c ˆ 0.86 in H2O; [a]D23
ˆ
40, c ˆ 0.5 in
Enzymatic oxidation of 2-deoxy-d-arabino-hexose (24):
H2O;[11b] [a]2D0
ˆ
13, c ˆ 0.5 in H2O[40]); MS m/z (%): 163 (100) [M ‡H],
‡
a) 2-Deoxy-a/b-d-erythro-hexos-3-ulose (27): Treatment of 24 (150 mg,
0.91 mmol) with PROD (32 units) and catalase (20000 U) for 6 d as
described in GP I gave a material consisting of 27 (80%) and 24 (20%)
(NMR). The crude product was separated by column chromatography
(CH2Cl2/MeOH 15:1), and filtered over celite yielding 27 (117 mg, 77%) as
a colourless syrup. [a]2D0 ˆ ‡33.9 (c ˆ 0.97 in H2O); MS m/z (%): 163 (6)
‡
145 (22) [M ‡H H2O]; C6H10O5 ´ 0.5H2O (171.2): calcd C 42.11, H 6.48;
found C 42.22, H 6.54.
b) 1,5-Anhydro-2-deoxy-2-methoximino-d-arabino-hexitol (47) and 1,5-
anhydro-2,3-dideoxy-2,3-bis(methoximino)-d-erythro-hexitol (48): Treat-
ment of 1,5-anhydro-d-glucitol (12, 100 mg, 0.61 mmol) with PROD
(70 units) for 3 d as described in GP I gave a mixture consisting of 20[40b]
and 1,5-anhydro-d-erythro-hex-2,3-diulose (21, NMR in Tables 4 and 5) in a
ratio of 2:1 (NMR). The mixture was treated without further purification in
MeOH (7 mL) with CH3ONH2 ´ HCl (220 mg, 2.67 mmol) and NaHCO3
(220 mg, 2.62 mmol) and stirred for 1 h. Evaporation and separation by
column chromatography (CH2Cl2/EtOAc 1:1) afforded 48 (40 mg, 30%) as
a colourless syrup, which was a mixture of four methoximino cis/trans
stereoisomers in a ratio of 66:30:3:1 (NMR). MS m/z (%): 219 (100)
‡
‡
‡
[M ‡H], 145 (100) [M ‡H H2O], 135 (38) [M ‡H CO], 117 (56)
‡
[M ‡H H2O CO]; NMR of the predominant isomer 2-deoxy-a-d-
erythro-hexos-3-ulopyranose: 1H NMR (D2O): d ˆ 2.401 (dd, J1,2e ˆ 1.2,
J2e,2a ˆ 14.4 Hz, 1H, H-2e), 2.812 (ddd, J2a,4 ˆ 1.1, J1,2a ˆ 4.4 Hz, 1H, H-2a),
3.683 (dd, J5,6b ˆ 3.8, J6a,6b ˆ 12.4 Hz, 1H, H-6b), 3.717 (dd, J5,6a ˆ 2.9 Hz,
1H, H-6a), 3.862 (ddd, J4,5 ˆ 10.1 Hz, 1H, H-5), 4.175 (dd, 1H, H-4), 5.530
(dd, 1H, H-1); 13C NMR (D2O): m ˆ 46.43 (C-2), 60.98 (C-6), 72.87 (C-4),
74.84 (C-5), 93.22 (C-1), 209.56 (C-3); C6H10O5 ´ 0.5H2O (171.2): calcd C
42.11, H 6.48; found C 42.37, H 5.89.
‡
‡
[M ‡H], 156 (25) [M
2MeO]; 1H NMR (D2O): d ˆ 3.840, 3.781 (2s, 6H,
OMe, first isomer 48a); 3.796, 3.791 (2s, 6H, OMe, second isomer 48b);
13C NMR (D2O): d ˆ 63.07, 63.14 (OMe, first isomer 48a); 63.15, 62.95
(OMe, second isomer 48b); other NMR data in Tables 4 and 5; C8H14N2O5
(218.1): calcd C 44.02, H 6.47, N 12.84; found C 43.87, H 6.67, N 12.71.
b) 1-Deoxy-a/b-d-ribulose (28): Treatment of 24 (200 mg, 1.22 mmol) with
PROD (32 units) and the normal amount of catalase for 8 d according to
GP I gave a material consisting of 65% of 28 and 58[55] and 35% of 24
(NMR). The crude product was separated by column chromatography
(CH2Cl2/MeOH 10:1), and 28 (88 mg, 54%) was obtained as a yellowish
syrup, which consisted of 28 and 58 in a ratio of 16:1. [a]2D0 of the mixture ˆ
Further elution of the column with CH2Cl2/MeOH (15:1) afforded
crystalline 47 (67 mg, 58%) as a single stereoisomer. M.p. 1098C; [a]D20
ˆ
24.5 (c ˆ 0.73 in H2O; ref. [53]: [a]D20
ˆ
37, c ˆ 1.3 in H2O, ref. [62]:
‡
4.5 (c ˆ 0.92 in H2O); MS m/z (%): 192 (100) [M ‡H], 174 (10)
[a]2D0
ˆ
19, c ˆ 1.4 in H2O; ref. [63]: [a]2D0 ˆ ‡28, c ˆ 1.3 in H2O); MS m/z
‡
[M ‡H H2O]; 1H NMR (D2O): d ˆ 3.687 (s, 3H, OMe); 13C NMR
‡
‡
‡
(%): 135 (4) [M ‡H], 117 (100) [M ‡H H2O], 99 (10) [M ‡H 2H2O];
NMR in Tables 6 and 7; C5H10O4 (134.1): calcd C 44.76, H 7.52; found C
44.26, H 7.34.
(D2O): d ˆ 61.74 (OMe); other NMR data in Tables 4 and 5; C7H13NO5
(191.1): calcd C 43.96, H 6.86, N 7.33; found C 43.83, H 7.10, N 6.79.
Table 4. 1H NMR data of 21, 47, and 48 in D2O.
Table 6. 1H NMR data of 28 and 58[a] in D2O.
21
47
48a
4.580
17.9
4.413
±
±
4.523
9.2
3.277
2.6
48b
4.672
17.2
4.243
±
±
4.189
7.0
3.451
2.9
28a
28b
28c
58
2.100
4.219
2.0
H-1a
J1a,1b
H-1b
H-3
J3,4
3.454
11.9
3.418
±
±
4.842
14.8
3.748
4.127
7.8
Me
2.094
1.266
1.246
3.734
5.0
H-3
J3,4
4.135
4.9
3.696
5.0
H-4
J4,5a
3.850
5.4
4.117
5.2
4.378
6.1
3.996
5.8
H-4
J4,5
3.445
9.9
3.349
9.2
J4,5b
H-5a
J5a,5b
H-5b
6.4
3.456
11.7
3.421
2.5
3.889
10.3
3.686
4.9
7.1
3.943
9.6
3.520
3.527
11.4
3.470
H-5
J5,6a
3.315
2.1
3.309
2.3
J5,6b
6.3
6.1
6.5
6.5
H-6a
J6a,6b
H-6b
3.670
12.5
3.477
3.696
12.3
3.519
3.695
12.5
3.492
3.659
12.2
3.530
[a] For 13C NMR see ref. [55].
Table 7. 13C NMR data of 28.
28a
28b
23.48
102.50
74.65
70.09
71.39
28c
Table 5. 13C NMR data of 21, 47 and 48 in D2O.
C-1
C-2
C-3
C-4
C-5
26.94
213.48
78.04
72.38
61.44
21.65
106.34
76.57
70.81
70.35
21
47
48a
48b
C-1
C-2
C-3
C-4
C-5
C-6
70.10[a]
93.38
94.56
69.59[a]
78.83
61.01
65.50
151.25
151.01
61.35
79.93
61.26
64.03
148.79
146.94
67.94
81.25
61.50
155.82
72.37
71.86
79.97
61.01
61.65
c) 2-Deoxy-d-arabino-hexonic acid characterised as 3,4,6-tri-O-acetyl-2-
deoxy-d-arabino-hexono-1,5-lactone (55), 3,4,5,6-tetra-O-acetyl-2-deoxy-d-
arabino-hexonic acid (56), and 3,4,6-tri-O-acetyl-2-deoxy-d-arabino-hexo-
no-1,4-lactone (57): Treatment of 24 (100 mg, 0.61 mmol) with PROD
(39 units) and the normal amount of catalase (presumably less active than
normal) for 6 d according to GP I gave a material (93 mg, 86%) consisting
of 2-deoxy-d-arabino-hexonic acid (29)[49, 50] and 2-deoxy-d-arabino-hex-
ono-1,5-lactone (53) in a ratio of 4:1 and traces of 2-deoxy-d-arabino-
hexono-1,4-lactone (54)[32, 50] (NMR, data in Tables 8 and 9). Purification on
a Dowex column gave a syrup (83 mg) consisting of calcium 2-deoxy-d-
arabino-hexonate (59), 2-deoxy-d-arabino-hexono-1,5 lactone (53), and
2-deoxy-d-arabino-hexono-1,4-lactone (54) in a ratio of 4:1:1 (NMR, data
in Tables 8 and 9). One drop of HClO4 (60%) was added to a stirred
[a] Signal assignment may be reversed.
a-d-Arabino-hexos-2-ulopyranosyl fluoride (22): Treatment of a-d-gluco-
pyranosyl fluoride (13, 300 mg, 1.65 mmol) with PROD (40 units) for 2 d as
described in GP I gave a colourless foam (350 mg) consisting of 22, a/b-d-
arabino-hexos-2-ulose (2) and 13 in a ratio of 4:3:3. Purification of all of 22
was not possible due to hydrolysis. However, by fast HPLC with small
amounts, 22 could be purified to about 90%, which allowed the assignment
of NMR signals. 1H NMR (D2O, hydrated form): d ˆ 3.404 (dd, J3,4 ˆ 9.5,
J4,5 ˆ 10.1 Hz, 1H, H-4), 3.567 (dd, J5,6b ˆ 5.7, J6a,6b ˆ 12.2 Hz, 1H, H-6b),
2450
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0412-2450 $ 17.50+.50/0
Chem. Eur. J. 1998, 4, No. 12

Pyranose 2-Oxidase
2442±2455
suspension of this syrup in Ac2O. After 2 h workup was performed as
described in GP II, the crude product was separated by HPLC on a silica gel
column (EtOAc/petroleum ether 1:1) to yield 56 (6 mg, 3%) as a colourless
syrup. [a]2D0 ˆ ‡30.2 (c ˆ 0.08 in CH2Cl2; ref. [52]: [a]2D0 ˆ ‡37.1, c ˆ 2.8 in
(0.1n), the solvent removed, and the residue filtered over a short silica gel
column (EtOAc/MeOH 5:1) giving a mixture (10 mg, 71%) of methyl b-d-
allopyranoside[66] and 25[57c, 67] in a ratio of 3:2 (NMR and TLC of authentic
sample).
‡
CHCl3); MS m/z (%): 303 (100) [M ‡H HOAc]; NMR in Tables 8 and 9.
Methyl b-d-xylo-hexos-3-ulo-pyranoside (31): Treatment of methyl b-d-
galactopyranoside (26, 300 mg, 1.54 mmol) with PROD (10 units) for 18 d
as described in GP I gave a material consisting of 26 (82%) and 31 (18%)
(NMR). The crude product was separated by column chromatography
Table 8. 1H NMR data of 29, 53, 54, 56 and 59 in D2O.
56[a,b]
59
2.338
15.1
8.3
(EtOAc/MeOH 8:1) yielding 31[23, 30] (26 mg, 9%) as a colourless syrup.
29
53
54
‡
[a]2D0
ˆ
24.3 (c ˆ 0.325 in H2O); MS m/z (%): 193 (6) [M ‡H], 175 (20)
H-2a
J2a,2b
J2a,3
2.490
15.6
8.5
2.911
17.5
6.0
2.875
17.8
5.6
2.581
16.0
7.6
‡
‡
‡
[M ‡H H2O], 161 (100) [M ‡H MeOH], 143 (44) [M ‡H H2O
‡
‡
MeOH], 115 (53) [M ‡H H2O MeOH CO], 97 (24) [M ‡H
2H2O MeOH CO]; 1H NMR (D2O): d ˆ 3.348 (d, J1,2 ˆ 8.1 Hz, 1H,
H-2b
J2b,3
2.441
5.1
2.422
8.1
2.397
0.4
2.527
6.3
2.285
5.3
H-2), 3.367 (s, 3H, OMe), 3.434 (d, J4,5 ˆ 1.0 Hz, 1H, H-4), 3.544 (dd, J5,6b
ˆ
5.2, J6a,6b ˆ 11.8 Hz, 1H, H-6b), 3.576 (dd, J5,6a ˆ 7.2 Hz, 1H, H-6a), 3.733
H-3
J3,4
4.128
2.0
3.907
7.5
4.528
3.5
5.550
2.5
4.058
2.0
(ddd, 1H, H-5), 4.224 (s, 1H, H-1).
Reduction of methyl b-d-xylo-hexos-3-ulo-pyranoside (31): Compound 31
(10 mg, 0.042 mmol) was treated as described for 30 giving a colourless
syrup (6 mg, 74%) consisting of methyl b-d-gulopyranoside[68] and
26[57d, 69]in a ratio of 55:45 (NMR). 1H NMR (D2O) of methyl b-d-
gulopyranoside: d ˆ 4.395 (d, J1,2 ˆ 8.8 Hz, 1H, H-1), 3.453 (s, 3H, OMe).
H-4
J4,5
H-5
J5,6a
3.284
8.4
3.547
3.0
3.575
9.3
4.040
2.4
4.8
3.739
4.297
9.2
3.845
2.8
5.4
3.660
5.361
8.9
5.154
2.6
3.264
8.3
3.541
2.9
6.4
3.639
J5,6b
6.2
4.6
H-6a
J6a,6b
H-6b
3.640
11.7
3.460
4.251
12.5
4.162
Hexyl 6-O-(b-d-glucopyranosyl)-b-d-glucopyranoside (36) and hexyl
2,3,4,2',3',4',6'-hepta-O-acetyl-6-O-(b-d-glucopyranosyl)-b-d-glucopyrano-
side (60): Treatment of hexyl b-d-glucopyranoside (32, 230 mg, 0.87 mmol)
with PROD (11 units) for 7 d as described in GP I gave a material
consisting of 32 (81%), a/b-d-arabino-hexos-2-ulose (2, 10%) and 36 (9%)
(NMR). The crude product was separated by column chromatography
(CH2Cl2/MeOH 3:1) giving 36 (30 mg, 8%) as a colourless syrup. The NMR
data (Table 10) were consistent with ref. [24]. Compound 36 was acetylated
as described in GP II. Separation by column chromatography (EtOAc/
petroleum ether 1:1) gave 60 (48 mg, 95%) as colourless crystals. M.p.
13.0
3.665
12.2
3.528
11.7
3.450
[a] In CDCl3. [b] Partly published.[53]
Table 9. 13C NMR data of 29, 53, 55, 56, 57 and 59 in D2O.
29
53
55[a]
56[a]
57[a]
59
C-1
C-2
C-3
C-4
C-5
C-6
176.36
38.93
67.16
72.82
71.30
63.30
173.88
36.72
68.18
68.75
81.51
60.49
167.01
33.73
68.65
67.80
76.22
62.18
n.d.
35.61[51]
67.15
70.00
68.23
61.80[51]
172.83
36.54
68.39
78.24
67.12
62.53
179.41
41.06
67.83
72.98
71.41
63.29
142 ± 1458C; [a]2D0
ˆ
21.2 (c ˆ 1.04 in CHCl3); MS m/z (%): 620 (15)
‡
‡
[M
OHex], 331 (100) [C14H19O9 ]; NMR in Tables 11 and 12; NMR of
the protecting groups: 1H NMR (CDCl3): d ˆ 1.979, 1.985, 2.009, 2.013,
2.016, 2.025, 2.079 (21H, CH3CO); 13C NMR (CDCl3): d ˆ 20.50 (2  ),
20.52, 20.54 (2 Â ), 20.57, 20.64 (CH3CO), 169.15, 169.23, 169.35, 169.57,
170.11, 170.17, 170.54 (CH3CO); C32H48O18 (720.3): calcd C 53.31, H 6.72;
found C 53.34, H 6.65.
[a] in CDCl3.
Phenyl 6-O-(b-d-glucopyranosyl)-b-d-glucopyranoside (37) and phenyl
2,3,4,2',3',4',6'-hepta-O-acetyl-6-O-(b-d-glucopyranosyl)-b-d-glucopyrano-
side (61): Treatment of phenyl b-d-glucopyranoside (33, 200 mg,
0.78 mmol) with PROD (24 units) for 3 d as described in GP I gave a
material consisting of 33 (68%), a/b-d-arabino-hexos-2-ulose (2, 25%) and
37 (7%) (NMR). The crude product was separated by preparative TLC
(acetone/BuOH/H2O 5:4:1) and filtered over celite giving 37[25] (21 mg,
6%) as a colourless syrup. Compound 37 was characterised by NMR
spectroscopy (Tables 10 and 13) and acetylated as described in GP II.
Separation by column chromatography (EtOAc/petroleum ether 1:1) gave
The second fraction (23 mg, 13%) consisted of 57 which crystallised from
EtOAc/petroleum ether. M.p. 1078C (ref. [61]: 103 ± 1058C); [a]2D0 ˆ ‡32.9
(c ˆ 0.05 in CH2Cl2); [a]2D0 ˆ ‡34 (c ˆ 1.3 in CHCl3); MS m/z (%): 289 (100)
‡
‡
[M ‡H], 229 (15) [M
HOAc]; NMR in Table 9.
The third fraction (11 mg, 6%) consisted of 55 as a colourless syrup. [a]D20
ˆ
‡63.2 (c ˆ 0.16 in CH2Cl2; ref. [64]: [a]2D0 ˆ ‡66.0, c ˆ 1.0 in EtOH); MS
‡
‡
m/z (%): 289 (100) [M ‡H], 247 (52) [M ‡H CH2CO], 229 (21)
‡
[M ‡H HOAc]; NMR in Table 9.
Methyl b-d-ribo-hexos-3-ulo-pyranoside (30): Treatment of methyl b-d-
glucopyranoside (25, 300 mg, 1.54 mmol) with PROD (28 units) for 8 d as
described in GP I gave a material consisting of 25 (86%) and 30 (14%)
(NMR). The crude product was separated by column chromatography
(EtOAc/MeOH 8:1) giving 30 (32 mg, 11%), which crystallised from
2-propanol. NMR in D2O showed a mixture of the keto isomer 30a and the
hydrated form 30b in a ratio of 2:1; 13C NMR data of 30a are not consistent
with ref. [30]. M.p. 1288C (ref. [65]: m.p. 130 ± 1328C from EtOAc/MeOH);
61 (26 mg, 73%) as colourless crystals. M.p. 1908C (ref. [25]: m.p. 194 ±
‡
1958C); [a]2D0
ˆ
30.1 (c ˆ 0.36 in CHCl3); MS m/z (%): 620 (4) [M
‡
‡
OPh], 391 (100) [C16H22O11], 331 (72) [C14H19O9 ]; NMR in Tables 11 and
12; NMR of the protecting groups: 1H NMR (CDCl3): d ˆ 1.851, 2.002,
2.017, 2.026, 2.051, 2.060, 2.083 (21H, CH3CO); 13C NMR (CDCl3): d ˆ
20.43 (2 Â ), 20.53 (2 Â ), 20.55, 20.58 (2 Â ) (CH3CO), 169.32, 169.37, 169.62,
170.13, 170.16, 170.54, 170.56 (CH3CO).
[a]2D0
ˆ
60.5 (c ˆ 0.68 in H2O; ref. [22]: [a]D20
ˆ
62, c ˆ 2.0 in H2O); MS
2-Nitrophenyl
2-nitrophenyl 3-O-(b-d-glucopyranosyl)-b-d-glucopyranoside (39), and
2-nitrophenyl 2,3,4,2',3',4',6'-hepta-O-acetyl-6-O-(b-d-glucopyranosyl)-
6-O-(b-d-glucopyranosyl)-b-d-glucopyranoside
(38),
‡
‡
m/z (%): 175 (5) [M ‡H H2O], 161 (100) [M ‡H MeOH], 115 (67)
‡
‡
[M ‡H H2O MeOH CO], 97 (14) [M ‡H 2H2O MeOH CO];
1
NMR of 30a: H NMR (D2O): d ˆ 3.358 (ddd, J4,5 ˆ 10.3, J5,6a ˆ 2.3, J5,6b
ˆ
b-d-glucopyranoside (62): Treatment of 2-nitrophenyl b-d-glucopyranoside
(34, 234 mg, 0.78 mmol) with PROD (7 units) for 3 d as described in GP I
gave a material consisting of 34 (83%), a/b-d-arabino-hexos-2-ulose (2,
2%), 38 (12%) and 39 (3%) (NMR). The crude product was separated by
preparative TLC (acetone/BuOH/H2O 5:4:1) and filtered over celite giving
39 (12 mg, 3%) and 38 (34 mg, 9%) as colourless residues. Compound 39
was identified by its Rf value which was identical to 41. Compound 38
(17 mg, 0.037 mmol) was characterised by NMR spectroscopy (Tables 10
and 13) and acetylated as described in GP II. Separation by column
chromatography (EtOAc/petroleum ether 1:1) gave 62 (26 mg, 94%) as
5.1 Hz, 1H, H-5), 3.434 (s, 3H, OMe), 3.654 (dd, J6a,6b ˆ 12.6 Hz, 1H, H-6b),
3.811 (dd, 1H, H-6a), 4.075 (dd, J1,2 ˆ 8.0 Hz, J2,4 ˆ 1.6, 1H, H-2), 4.177 (dd,
1H, H-4), 4.306 (d, 1H, H-1); 13C NMR (D2O): d ˆ 57.64 (OMe), 61.27 (C-
6), 72.35 (C-4), 76.20 (C-5), 76.85 (C-2), 104.67 (C-1), 206.68 (C-3);
1H NMR (D2O) of 30b: d ˆ 3.281 (d, J4,5 ˆ 10.1 Hz, 1H, H-4), 3.316 (d,
J1,2 ˆ 8.3 Hz, 1H, H-2), 3.365 (s, 3H, OMe), 3.384 (ddd, J5,6a ˆ 2.3, J5,6b
ˆ
6.0 Hz, 1H, H-5), 3.522 (dd, J6a,6b ˆ 12.5 Hz, 1H, H-6b), 3.711 (dd, 1H,
H-6a), 4.249 (d, 1H, H-1).
Reduction of methyl b-d-ribo-hexos-3-ulo-pyranoside (30): Compound 30
(14 mg, 0.073 mmol) in MeOH (2 mL) was treated with sodium borohy-
dride (11 mg, 0.40 mmol) for 30 min. The solution was neutralised with HCl
colourless crystals. M.p. 1958C; [a]D20
ˆ
10.3 (c ˆ 0.91 in CHCl3); MS m/z
‡
‡
(%): 620 (27) [M
OPhNO2], 331 (100) [C14H19O9 ]; NMR in Tables 11
Chem. Eur. J. 1998, 4, No. 12
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0412-2451 $ 17.50+.50/0
2451

S. Köpper et al.
FULL PAPER
Table 10. 1H NMR data of 23, 36, 37, 38, 40 and 65 in D2O.
23a
23b
23c
36
4.259[b]
37
4.991
±
±
7.6
3.401
9.2
38
5.077
±
±
7.2
3.449
9.3
40
5.093
±
±
7.2
3.450
9.2
65a
3.465
12.1
3.424
±
±
65b
H-1a
J1a,1b
H-1b
J1,2
H-2
J2,3
4.745
±
±
±
±
4.510
±
±
±
±
±
4.740
±
±
±
±
±
3.386
12.1
3.329
±
±
±
±
±
7.9[b]
3.122
9.3
±
±
H-3
J3,4
H-4
J4,5
H-5
J5,6a
J5,6b
H-6a
J6a,6b
H-6b
H-1'
J1',2'
H-2'
J2',3'
H-3'
J3',4'
3.556
9.3
3.393
10.1
3.842
1.9
3.343
9.4
3.330
9.4
3.437
1.9
6.2
4.004
3.995
7.8
3.899
8.3
3.273
9.9
3.297
9.0
3.413
1.6
5.7
3.448
9.2
3.373
9.4
3.646
1.6
6.0
3.426
8.8
3.370
9.4
3.650
1.6
6.1
3.379
9.2
3.450
n.d.
3.707
n.d.
ꢀ 5.5[a]
4.021
ꢀ 11[a]
3.709
4.275
8.0
3.098
9.3
3.237
9.3
3.902
5.2
3.818
7.1
3.988
2.2
7.8
3.945
8.1
3.918
6.5
3.761
2.7
7.2
3.925
11.2
3.586
4.325
7.9
3.118
9.4
3.300
9.4
3.753
n.d.
n.d.
n.d.
n.d.
n.d.
4.336
n.d.
3.123
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
5.5
3.958
11.6
3.698
4.309
7.9
3.123
9.4
3.310
9.4
3.193
9.4
3.270
2.0
3.995
11.8
4.014
12.0
3.998
11.9
3.942
11.4
11.5
3.648
4.319
8.0
3.669
4.322[b]
7.9
3.067
9.0
3.202
9.7
3.291
9.3
3.253
2.0
5.5
3.723
11.9
3.728
4.296
8.0
3.102
9.0
3.248
9.0
3.187
9.6
3.145
2.0
5.8
3.693
12.0
3.498
3.715
4.262
8.0
3.084
9.0
3.229
9.0
3.173
AB
3.165
1.4
5.3
3.691
3.601
4.312
7.9
3.113
9.3
3.304
9.3
3.188
9.3
3.266
2.2
6.1
3.716
12.5
3.518
3.123
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
H-4'
J4',5'
3.181
9.3
3.151
1.5
3.184
9.6
3.266
2.2
H-5'
J5',6'a
J5',6'b
H-6'a
J6'a,6'b
H-6'b
5.7
5.5
6.1
3.723
12.4
3.527
3.682
12.3
3.484
3.716
12.5
3.518
12.2
3.489
3.531
[a] ABX system. [b] In agreement with ref. [24].
Table 11. 1H NMR data of 49 ± 51 and 60 ± 64 in CDCl3.
49a
50a
51a
51b
60
61
62
63
64
H-1a
J1a,1b
J1,2
6.315
4.526
±
6.072
4.520
±
±
1.8
7.9
±
±
4.886
4.482
17.3
±
±
8.0
4.680
±
4.612
4.584
12.0
±
±
8.0
4.399
±
±
4.997
±
4.446
4.592
±
±
8.0
8.0
±
±
5.075
4.581
±
±
7.7
8.0
±
±
5.163
4.540
±
±
7.5
7.9
±
±
5.077
4.646
±
±
7.7
8.0
±
±
5.101
4.576
±
±
7.5
8.0
±
±
±
3.7
7.9
±
H-1b
H-2
J2,3
±
5.051
4.980
10.2
5.240
5.008
3.5
±
4.919
4.972
9.8
5.231
4.989
9.5
5.270
5.011
9.5
5.316
4.936
9.2
5.280
5.019
9.9
4.989
±
9.6
9.5
9.6
9.6
9.7
9.4
9.3
9.5
9.6
H-3
J3,4
5.450
5.199
9.6
5.331
5.206
10.0
5.484
5.193
2.1
5.767
5.217
3.9
5.173
5.164
9.1
5.290
5.134
9.2
5.319
5.170
9.0
4.028
5.167
9.2
5.297
5.169
9.1
9.4
9.4
9.7
9.6
9.2
9.4
9.3
9.4
9.4
H-4
J4,5
5.011
5.073
10.4
5.224
5.070
10.4
5.638
5.081
8.8
5.099
5.089
6.5
4.866
5.054
10.0
10.1
3.661
4.962
5.064
9.9
4.996
5.082
9.9
10.0
3.943
3.649
2.0
5.042
5.089
9.3
4.977
5.078
9.9
10.0
3.915
3.690
2.0
9.9
9.9
9.6
9.8
9.7
9.7
H-5
J5,6a
4.060
3.686
1.9
3.980
3.686
1.9
5.177
3.711
3.1
4.393
3.711
3.1
3.872
3.588
1.8
3.922
3.711
4.9
3.672
1.5
4.5
4.6
4.4
4.4
4.7
4.8
4.8
4.3
5.0
J5,6b
H-6a
J6a,6b
H-6b
5.7
2.0
6.2
2.0
5.1
2.4
5.6
2.4
7.4
1.8
8.1
2.2
7.6
2.0
3.4
2.4
n.d.
4.388
n.d.
12.4
n.d.
4.075
8.2
2.4
3.934
4.267
11.1
12.3
3.526
4.127
3.960
4.271
11.1
12.3
3.524
4.127
4.008
4.267
11.2
12.4
3.540
4.143
4.102
4.267
11.1
12.4
3.675
4.143
3.842
4.256
10.8
12.4
3.610
4.114
3.869
4.240
11.6
12.2
3.691
4.116
3.910
4.262
11.3
12.3
3.660
4.143
3.888
4.248
11.4
12.3
3.718
4.155
2452
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0412-2452 $ 17.50+.50/0
Chem. Eur. J. 1998, 4, No. 12

Pyranose 2-Oxidase
2442±2455
Table 12. 13C NMR data of 49 ± 51 and 60 ± 64 in CDCl3.
49a
50a
51a
51b
61.69
60
61
98.88
62
98.20
63
98.06
64
C-1/C-1'
C-2/C-2'
C-3/C-3'
C-4/C-4'
C-5/C-5'
C-6/C-6'
88.82
90.31
66.64
100.52
100.73
71.36[a]
71.09[a]
72.74[b]
72.78[b]
69.17
68.26
73.31
71.92
68.26
100.19
100.62
70.48
71.14
72.32
72.75
68.73
68.24
74.06
72.06
68.12
61.77
100.72
69.19
70.91
69.92
72.76
68.39
68.30
70.85
71.94
67.65
61.80
100.79
68.39[a]
70.91
68.81
72.63
65.83
68.30[a]
71.72
71.86
67.98
61.80
100.50
100.81
70.91
74.17
72.68
68.59[a]
68.37
68.68[a]
71.95
66.79
61.88
100.93
108.02
71.06
78.98
72.73
76.02
68.37
81.72
71.89
68.54
61.88
100.44
71.14
71.18
72.61
72.80
68.82
68.23
73.97
71.90
67.92
61.72
100.51
70.94
71.03
72.40
72.58
68.56
68.24
73.88
72.09
67.79
61.76
100.94
72.38
71.02
78.39
72.81
67.96
68.13
72.15
71.77
62.01
61.63
61.78
[a] Signal assignment may be reversed. [b] Signal assignment may be reversed.
Table 13. 13C NMR data of 37, 38, 40, 41 and 65 in D2O.
(1S,2R,3R,4R,6S)-3,6-dihydroxy-1,2-O-isopropylidene-5,7-dioxabicy-
clo[2,2,2]octan (42a), and (1S,2R,3R,4R,6R)-3,6-dihydroxy-1,2-O-isopro-
pylidene-5,7-dioxabicyclo[2,2,2]octan (42b), and (2S)-1,2:2,3:5,6-tri-O-iso-
propylidene-b-d-ribo-hexos-2-ulo-1,4-furanose (44): Isopropylidenation of
crude 15 (120 mg, 0.61 mmol) as described in GP III and separation by
37
38
40
41
65a
65b
62.90
C-1
C-2
C-3
C-4
C-5
C-6
C-1'
C-2'
C-3'
C-4'
C-5'
C-6'
100.25
73.18
75.77[a]
69.64
75.63
68.53
102.80
73.41
75.89[a]
69.94
76.12
61.01
100.71
72.90
75.86[a]
69.40
75.65
68.62
102.96
73.36
75.89[a]
69.90
76.12
60.97
99.57
99.50
63.05
72.95
75.66
69.45
75.54
68.65
102.98
73.35
75.92
69.89
76.12
60.96
72.78
84.14
68.09
76.27
60.67
103.06
73.72
75.89
69.85
76.16
60.97
n.d.
82.05
102.07
74.98
75.38
79.51
71.28
102.91
73.42
75.84
69.89
76.22
61.00
column chromatography (n-hexane/acetone 5:1) gave 44 (8 mg, 4%). M.p.
‡
838C; [a]2D0
ˆ
29.2 (c ˆ 0.04 in CH2Cl2); MS m/z (%): 259 (100) [M ‡H
76.56
80.20
71.28
102.91
73.42
75.84
69.97
76.22
61.00
(CH3)2CO]; NMR in Tables 14 and 15; NMR of the protecting groups:
1H NMR (CDCl3): d ˆ 1.371, 1.442, 1.468, 1.504, 1.532, 1.546 (18H, CH3);
13C NMR (CDCl3): d ˆ 25.16, 25.35, 26.78, 26.92, 27.10, 27.31 (CH3), 113.86,
114.77, 120.16 ((CH3)2C); C15H24O7 (316.2): calcd C 56.95, H 7.65; found C
56.65, H 7.58.
Table 14. 1H NMR data of 42 ± 46 in CDCl3.
42a
42b
43a
43b
44
45[a]
46
[a] Signal assignment may be reversed.
H-1
H-3
J3,4
H-4
J4,5
H-5
J5,6a
J5,6b
H-6a
J6a,6b
H-6b
4.829
4.771
5.8
4.490
0.0
4.178
2.1
0.0
4.941
4.754
5.9
4.667
0.0
4.121
1.9
0.0
6.280
4.742
5.7
5.159
0.0
4.327
1.7
0.0
6.173
4.450
5.7
5.072
0.0
4.309
1.3
0.0
5.534
4.591
0.0
4.124
9.7
4.238
5.8
4.8
5.586
4.598
4.5
4.070
4.8
4.178
6.8
6.7
5.404
4.159
6.1
3.421
9.7
3.709
6.1
±
and 12; NMR of the protecting groups: 1H NMR (CDCl3): d ˆ 1.883, 2.015,
2.034, 2.036, 2.070, 2.091, 2.115 (21H, CH3CO); 13C NMR (CDCl3): d ˆ
20.50, 20.55 (4 Â ), 20.60, 20.70 (CH3CO); C32H39NO20 (757.7): calcd C
50.73, H 5.19, N 1.85; found C 50.31, H 5.16, N 1.81.
4-Nitrophenyl 6-O-(b-d-glucopyranosyl)-b-d-glucopyranoside (40), 4-ni-
trophenyl 3-O-(b-d-glucopyranosyl)-b-d-glucopyranoside (41), 4-nitro-
phenyl 2,4,6,2',3',4',6'-hepta-O-acetyl-3-O-(b-d-glucopyranosyl)-b-d-glu-
copyranoside (63), and 4-nitrophenyl 2,3,4,2',3',4',6'-hepta-O-acetyl-6-O-
(b-d-glucopyranosyl)-b-d-glucopyranoside (64): Treatment of 4-nitrophen-
yl b-d-glucopyranoside (35, 222 mg, 0.74 mmol) with PROD (7 units) for
3 d as described in GP I gave a material consisting of 35 (63%), a/b-d-
arabino-hexos-2-ulose (2, 13%), 40 (16%), and 41 (8%) (NMR). The
crude product was separated by preparative TLC (acetone/BuOH/H2O
5:4:1) and filtered over celite giving 41 (16 mg, 5%) and 40[26] (26 mg, 8%)
as colourless residues. Both compounds were characterised by NMR
4.005
12.0
3.364
3.823
12.0
3.707
4.053
11.7
3.459
3.916
11.9
3.724
4.134
8.9
3.965
3.885
8.3
3.754
1.342
±
±
[a] In C6D6.
Table 15. 13C NMR data of 42 ± 46 in CDCl3.
42a 42b 43a 43b
91.69 94.23 88.48
100.87 101.81 103.82 103.46 110.00 109.95 107.45
44
45[a]
46
spectroscopy (Tables 10 and 13). 1H NMR (D2O) of 41: d ˆ 5.206 (d, J1,2
ˆ
C-1
C-2
C-3
C-4
C-5
C-6
90.89 108.32 107.79 102.50
7.4, 1H, H-1), 4.697 (d, J1,2 ˆ 8.0 Hz,1H, H-1').
An analytical amount of 41 was acetylated as described in GP II. Treatment
of the crude product with EtOAc/petroleum ether lead to crystals of 63,
which were filtered and measured by NMR spectroscopy (Tables 11 and
12). NMR of the protecting groups: 1H NMR (CDCl3): d ˆ 2.002, 2.051,
2.077 (2  ), 2.096, 2.163 (21H, CH3CO); 13C NMR (CDCl3): d ˆ 20.36,
20.45, 20.53 (2 Â ), 20.65, 20.71, 20.85 (CH3CO); 168.70, 169.16, 169.25,
169.34, 170.35, 170.46 (2 Â ) (CH3CO).
81.50
78.62
79.23
59.79
82.12
76.43
78.05
66.84
81.20
79.25
79.76
61.81
81.33
78.53
80.02
67.01
84.44
87.44
74.92
67.27
83.65
85.25
75.78
65.71
85.07
75.30
66.85
17.03
[a] In C6D6.
Further elution with n-hexane/acetone 2:1 gave 42 (52 mg, 38%) as a
colourless syrup. H NMR showed a a/b ratio of 4:1. After crystallisation
Compound 40 (40 mg, 0.13 mmol) was acetylated as described in GP II.
Separation by column chromatography (EtOAc/petroleum ether 1:1)
afforded 64 (52 mg, 80%) as colourless crystals. M.p. 2158C (ref. [26]:
1
from 2-propanol the a/b ratio was 11:1. M.p. 1508C; [a]D20
ˆ
3.6 (c ˆ 0.03
‡
‡
m.p. 215 ± 2168C); [a]D20
ˆ
46.0 (c ˆ 1.25 in CHCl3); ref. [26]: [a]D20
ˆ
in CH2Cl2); MS m/z (%): 219 (72) [M ‡H], 201 (100) [M ‡H H2O];
NMR in Tables 14 and 15; NMR of the protecting group for the a-anomer:
1H NMR (CDCl3): d ˆ 1.414, 1.543 (6H, CH3); 13C NMR (CDCl3): d
ˆ25.05, 26.04 (CH3), 113.29 ((CH3)2C); for the b-anomer: 1H NMR
(CDCl3): d ˆ 1.405, 1.555 (6H, CH3); 13C NMR (CDCl3): d ˆ 24.99, 26.01
(CH3), 112.84 ((CH3)2C); C9H14O6 (218.1): calcd C 49.52, H 6.47; found C
49.23, H 6.55.
‡
‡
47.35 in CHCl3; MS m/z (%): 759 (18) [M ‡H], 620 (100) [M
OPh-
‡
NO2], 331 (86) [C14H19O9 ]; NMR in Tables 11 and 12; NMR of the
1
protecting groups: H NMR (CDCl3): d ˆ 1.872, 2.041, 2.048, 2.060, 2.082,
2.102, 2.112 (21H, CH3CO); 13C NMR (CDCl3): d ˆ 20.45, 20.53 (2  ),
20.55 (2 Â ), 20.58, 20.69 (CH3CO), 169.12, 169.19, 169.37, 169.57, 170.04,
170.14, 170.50 (CH3CO).
Chem. Eur. J. 1998, 4, No. 12
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0412-2453 $ 17.50+.50/0
2453

S. Köpper et al.
FULL PAPER
(1R,2R,3R,4R,6R)-3,6-diacetoxy-1,2-O-isopropylidene-5,7-dioxabicy-
clo[2,2,2]octan (43a) and (1R,2R,3R,4R,6S)-3,6-diacetoxy-1,2-O-isopro-
pylidene-5,7-dioxabicyclo[2,2,2]octan (43b): Acetylation of 42 (19 mg,
0.09 mmol) as described in GP II and separation by column chromatog-
raphy (n-hexane/acetone 5:1) gave a colourless syrup (25 mg, 95%)
consisting of 43a and 43b in a ratio of 58:42 (NMR). After crystallisation
from 2-propanol the ratio was 44:56. M.p. 838C; [a]2D0 ˆ ‡1.1 (c ˆ 0.26 in
Table 16. 13C NMR data of 6-O-(b-d-glucopyranosyl)-d-arabino-hexos-2-
ulose (23) in D2O.
23a
23b
23c
C-1/C-1'
C-2/C-2'
C-3/C-3'
C-4/C-4'
C-5/C-5'
C-6/C-6'
94.80
95.24
90.18
102.88
93.70
73.30
73.67
75.80
68.77
69.85
71.14
76.09
69.08
60.95
102.88
93.13
73.30
76.42
75.80
68.63
69.85
75.04
76.09
69.05
60.95
102.88
101.24
73.30
75.80
75.80
75.04
69.85
81.17
76.09
n.d.
‡
‡
CH2Cl2); MS m/z (%): 243 (59) [M ‡H HOAc], 215 (100) [M ‡H
CH3CHO CO2], 201 (18); NMR in Tables 14 and 15; NMR of the
protecting groups for the a-anomer: 1H NMR (CDCl3): d ˆ 1.370, 1.479
(6H, (CH3)2C), 2.109, 2.161 (6H, CH3CO); 13C NMR (CDCl3): d ˆ 20.68,
20.86 (CH3CO), 25.47, 26.02 ((CH3)2C), 113.34 ((CH3)2C), 167.78, 169.39
(CH3CO); for the b-anomer: 1H NMR (CDCl3): d ˆ 1.383, 1.505 (6H,
(CH3)2C), 2.096, 2.101 (6H, CH3CO); 13C NMR (CDCl3): d ˆ 20.77, 21.27
(CH3CO), 25.04, 25.95 ((CH3)2C), 112.86 ((CH3)2C), 167.30, 168.54
(CH3CO); C13H18O8 (302.3): calcd C 51.65, H 6.00; found C 51.21, H 6.08.
60.95
(2R)-1,2:2,3:5,6-Tri-O-isopropylidene-a-d-lyxo-hexos-2-ulo-1,4-furanose
(45): Isopropylidenation of 16 (435 mg, 2.22 mmol) as described in GP III
and separation by HPLC on a silica column (n-hexane/acetone 2:1) gave 45
Acknowledgments: This work was supported by the Bundesminsterium für
Forschung und Technologie (Grant No. 0319515A and 0319516A).
(23 mg, 3%). M.p. 528C; [a]D20
ˆ
23.2 (c ˆ 0.24 in CH2Cl2); MS m/z (%):
‡
‡
259 (100) [M ‡H (CH3)2CO], 201 (50) [M ‡H 2(CH3)2CO]; NMR in
Tables 14 and 15; NMR of the protecting groups: 1H NMR (C6D6): d ˆ
1.273, 1.382, 1.394, 1.406, 1.438, 1.504 (18H, CH3); 13C NMR (C6D6): d ˆ
25.63, 25.80, 26.69, 27.03, 28.03, 28.08 (CH3), 116.16, 116.66, 121.22
((CH3)2C); C15H24O7 (316.2): calcd C 56.95, H 7.65; found C 57.03, H 7.86.
Received: April 21, 1998 [F1108]
[1] H. W. Ruelius, R. M. Kerwin, F. W. Janssen, Biochim. Biophys. Acta
1968, 167, 493 ± 500 and 501 ± 510.
[2] A. Schäfer, S. Bieg, A. Huwig, G.-W. Kohring, F. Giffhorn, Appl.
Environ. Microbiol. 1996, 63, 2586 ± 2592.
[3] G. Daniel, J. Volc, E. Kubatova, Appl. Environ. Microbiol. 1994, 60,
2524 ± 2532.
(2R)-6-Deoxy-1,2:2,3-di-O-isopropylidene-a-d-arabino-hexos-2-ulo-1,5-p-
yranose (46): Isopropylidenation of 19 (49 mg, 0.27 mmol) as described in
GP III and separation by HPLC on a silica column (n-hexane/acetone 2:1)
gave 46 (7 mg, 10%). M.p. 588C; [a]D20
ˆ
43.6 (c ˆ 0.11 in CH2Cl2); MS m/
‡
z (%): 130 (100) [M ‡H 2(CH3)2CO CH3], 119 (29), 110 (37); NMR in
1
Tables 14 and 15; NMR of the protecting groups: H NMR (CDCl3): d ˆ
[4] Y. Machida, T. Nakanishi, Agric. Biol. Chem. 1990, 54, 799 ± 801.
[5] a) H. J. Danneel, F. Giffhorn in DECHEMA Biotechnology Confer-
ences, Vol. 4, Part A (Eds.: D. Behrens, P. Krämer), VCH, Weinheim,
1990, pp. 119 ± 122; b) A. Huwig, A. Brans, H. J. Danneel, S. Köpper,
F. Giffhorn, Med. Fac. Landbouw. Univ. Gent 1992, 57, 1749 ± 53;
c) H. J. Danneel, E. Rössner, A. Zeeck, F. Giffhorn, Eur. J. Biochem.
1993, 214, 795 ± 802.
[6] K. E. Koths, R. F. Halenbeck, D. B. Ring, US4568645 1986 and Chem.
Abstr. 1986, 104, 166931v; K. E. Koths, R. F. Halenbeck, P. M.
Fernandez, US4569910 1986 and Chem. Abstr. 1986, 104, 184888r;
K. E. Koths, R. F. Halenbeck, US4569913 1986 and Chem. Abstr.
1986, 104, 184878n; I.-N. Liu, B. Wolf, J. Geigert, S. L. Neidleman,
J. D. Chin, D. S. Hirano, Carbohydr. Res. 1983, 113, 151 ± 157; J.
Geigert, S. L. Neidleman, D. S. Hirano, Carbohydr. Res. 1983, 113,
159 ± 162.
[7] a) J. Volc, P. Sedmera, V. Musilek, Folia Microbiol. (Prague) 1978, 23,
292 ± 298; b) J. Volc, P. Sedmera, V. Musilek, Coll. Czech. Chem. 1980,
45, 950 ± 955; c) J. Volc, P. Sedmera, V. Musilek, CS175897 1979 and
Chem. Abstr. 1979, 90, 136263d.
[8] J. Gabriel, J. Volc, P. Sedmera, G. Daniel, E. Kubatova, Arch.
Microbiol. 1993, 160, 27 ± 34.
[9] S. P. Wolff, R. T. Dean, Biochem. J. 1987, 245, 243 ± 250.
[10] S. Kawashiki, J. Tsunehiro, K. Uchida, Carbohydr. Res. 1991, 211,
167 ± 171.
[11] K. J. Knecht, M. S. Feather, J. W. Baynes, Arch. Biochim. Biophys.
1992, 294, 130 ± 137.
[12] H. Yamada, S. Miyata, N. Igaki, H. Yatabe, Y. Miyauchi, T. Ohara, M.
Sakai, H. Shoda, M. Oimomi, M. Kasuga, J. Biol. Chem. 1994, 269,
20275 ± 20280.
[13] a) T. Nakamura, A. Naito, Y. Takahashi, H. Akanuma, J. Biochem.
1986, 99, 607 ± 613; b) G. Deffieux, R. Baute, M.-A. Baute, M. Atfani,
A. Carpy, Phytochemistry 1987, 26, 1391 ± 1393.
[14] F. W. Lichtenthaler, S. Schwidetzky, K. Nakamura, Tetrahedron Lett.
1990, 31, 71 ± 74.
[15] S. Freimund, A. Huwig, F. Giffhorn, S. Köpper, J. Carbohydr. Chem.
1996, 15, 115 ± 120.
[16] a) J. Volc, E. Kubatova, P. Sedmera, G. Daniel, J. Gabriel, Arch.
Microbiol. 1991, 156, 297 ± 301; b) K. Koths, R. Halenbeck, M.
Moreland, Carbohydr. Res. 1992, 232, 59 ± 75; c) J. Gabriel, J. Volc, E.
Kubatova, Z. Palkova, M. Pospisek, Carbohydr. Res. 1994, 252, 297 ±
301.
1.472, 1.483, 1.495, 1.505 (12H, CH3); 13C NMR (CDCl3): d ˆ 25.50, 27.02,
27.35, 27.39 (CH3), 109.63, 110.79 ((CH3)2C); C12H20O6 (260.1): calcd C
55.36, H 7.75; found C 55.14, H 8.03.
1,5-Anhydro-2-deoxy-2-methoximino-d-arabino-hexitol (47): Treatment
of pure 34 (50 mg, 0.29 mmol) with CH3ONH2 ´ HCl (80 mg, 0.96 mmol)
and NaHCO3 (80 mg, 0.95 mmol) as described for the preparation of 48 and
47 gave, after separation by column chromatography (CH2Cl2/MeOH
15:1), pure 47 (50 mg, 90%) as a single isomer.
1,2,3,4,2',3',4',6'-Octa-O-acetyl-6-O-(b-d-glucopyranosyl)-a- (49a) and -b-
d-glucose (49b), 1,2,3,4,2',3',4',6'-octa-O-acetyl-6-O-(b-d-glucopyranosyl)-
a- (50a) and -b-d-mannose (50b), 1,2,3,4,2',3',4',6'-octa-O-acetyl-6-O-(b-d-
glucopyranosyl)-a- (51a) and -b-d-fructofuranose (51b), and 6-O-(b-d-
glucopyranosyl)-a/b-d-fructofuranose (65): Treatment of 6-O-(b-d-gluco-
pyranosyl)-d-glucose (14, 200 mg, 0.58 mmol) with PROD (27 units) for 3 d
as described in GP I gave a material consisting of 6-O-(b-d-glucopyrano-
syl)-a/b-d-arabino-hexos-2-ulose (23, 80%), 14 (10%) and a/b-d-arabino-
hexos-2-ulose (2, 10%, NMR). 1H NMR (D2O) of 23b: 3.123 (dd, J1,2 ˆ 8.0,
1H, H-2'), 3.330 (dd, J3,4 ˆ 9.4, J4,5 ˆ 9.4, 1H, H-4), 3.343 (d, 1H, H-3), 3.437
(ddd, J5,6a ˆ 1.9, J5,6b ˆ 6.2, 1H, H-5), 3.648 (dd, J6a,6b ˆ 11.5 Hz, 1H, H-6b),
4.004 (dd, 1H, H-6a), 4.319 (d, 1H, H-1'), 4.510 (s, 1H, H-1); 1H NMR
(D2O) of 23c: 3.123 (dd, J1,2 ˆ 8.0, 1H, H-2'), 3.753 (ddd, J4,5 ˆ 8.3, 1H,
H-5), 3.899 (dd, J3,4 ˆ 7.8 Hz, 1H, H-4), 3.995 (d, 1H, H-3), 4.336 (d, 1H,
H-1'), 4.740 (s, 1H, H-1); other NMR data in Tables 10 and 16.
The crude product was hydrogenated for 10 days as described in GP IVand
immediately acetylated as described in GP II. Separation by column
chromatography (EtOAc/petroleum ether 2:3) gave a mixture (258 mg) of
three peracetylated disaccharides in a ratio of 1:1:1 (NMR). MS m/z (%):
‡
‡
679 (2) [M ‡H], 619 (100), 331 (100) [M ‡H HOAc], 331 (41)
‡
[C14H19O9 ]; 49: a/b ratio 1:1; 50: a/b ratio 3:1; 51: a/b ratio 1:6; NMR in
1
Tables 11 and 12; H NMR signal (CDCl3) of 50b: 5.832 (d, J1,2 ˆ 1.1 Hz,
1H, H-1); C28H38O19 (678.6): calcd C 49.56, H 5.64; found C 49.39,
H 5.69.
Further elution of the column afforded a mixture of two partially acetylated
disaccharides. After dissolving in MeOH (3 mL), adding NaOMe (30 mL of
a 1.5m solution in MeOH), and stirring for 1 h, the solution was neutralised
‡
with DOWEX 50W-X8 (H ) and filtered. HPLC (Merck LiChrosorb RP18,
7 mm, 250 Â 10 mm, 1 mLmin 1 H2O) gave 65 (4 mg) as a colourless syrup.
NMR (Tables 10 and 13) showed a a/b ratio of 1:4.
2454
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0412-2454 $ 17.50+.50/0
Chem. Eur. J. 1998, 4, No. 12

Pyranose 2-Oxidase
2442±2455
[17] E. W. Baxter, A. B. Reitz, Bioorg. Med. Chem. Lett. 1992, 2, 1419 ±
1422; E. W. Baxter, A. B. Reitz, Tetrahedron Lett. 1990, 31, 6777 ±
6780.
[18] A. Huwig, H. J. Danneel, F. Giffhorn, J. Biotechnol. 1994, 32, 309 ±
315.
[19] O. Theander, Acta Chem. Scand. 1958, 12, 1887 ± 1896.
[20] F. W. Lichtenthaler, E. S. H. El Ashry, V. H. Göckel, Tetrahedron Lett.
1980, 21, 1429 ± 1432.
[21] Unpublished results; NMR data of 1,5-anhydro-d-tagatose and the
2,3-diketo compound are available.
[47] We synthesised the glycoside 52 by glycosylation of 3,4,6-tri-O-acetyl-
2-bromo-2-deoxy-a-d-glucopyranosyl bromide (A. Fogh, I. Lundt, C.
Pedersen, P. Rasmussen, Acta Chem. Scand. Ser. B 1977, 31, 768 ± 770)
with 4-methoxybenzyl alcohol (yield 46%), followed by hydrogena-
tion/deacetylation (52%) and oxidation with bis(tributyltin)oxide/Br2
(ref. [28]), which selectively gave compound 52 in a yield of 60%.
Upon treatment of 52 with HCl at pH 4 and 65oC we observed
formation of the keto derivative 27 (50%), 1-deoxy-d-ribulose (28,
10%) and unidentified products (40%) NMR spectroscopically. The
amount of 28 was raised to 40% when air was bubbled through the
mixture during the reaction.
[22] B. Lindberg, O. Theander, Acta Chem. Scand. 1954, 8, 1870 ± 1874.
[23] E. Brimacombe, J. S. Brimacombe, B. Lindberg, O. Theander, Acta
Chem. Scand. 1961, 15, 437 ± 438.
[48] J.-C. Fischer, D. Horton, W. Weckerle, Can. J. Chem. 1977, 55, 4078 ±
4089.
[24] M. Yuda, K. Ohtani, K. Mizutani, R. Kasai, O. Tanaka, M.-R. Jia, Y.-
R. Ling, X.-F. Pu, Y.-I. Saruwatari, Phytochemistry 1990, 29, 1989 ±
1993.
[25] B. Helferich, E. Schmitz-Hillebrecht, Chem. Ber. 1933, 66, 378 ± 383.
[26] F. H. Babers, W. F. Goebel, J. Biol. Chem. 1934, 105, 473 ± 480.
[27] S. Freimund, S. Köpper, J. Carbohydr. Chem. in press.
[28] P. L. Barili, G. Berti, F. DꢁAndrea, V. Di Bussolo, A. Gaudiosi,
Tetrahedron Lett. 1992, 33, 7061 ± 7063.
[49] P. E. Morris Jr., K. D. Hope, D. E. Kiely, J. Carbohydr. Chem. 1989, 8,
515 ± 530.
[50] I. W. Hughes, W. G. Overend, M. Stacey, J. Chem. Soc. 1949, 2846 ±
2849; H. S. Isbell, H. L. Frush, R. Naves, P. Soontracharoen, Carbo-
hydr. Res. 1981, 90, 111 ± 122; K. Bock, I. Lundt, C. Pedersen, Acta
Chem. Scand. Ser. B 1984, 38, 555 ± 561.
[51] P. A. Levene, L. A. Mikeska, J. Biol. Chem. 1930, 88, 791 ± 798; M.
Rizzotto, S. Signorella, M. I. Frascaroli, V. Daier, L. F. Sala, J.
Carbohydr. Chem. 1995, 14, 45 ± 51.
[52] M. Gouedard, F. Gaudemer, A. Gaudemer, Bull. Soc. Chim. Fr. 1973,
40, 577 ± 580.
[29] P. L. Barili, G. Berti, G. Catelani, F. DꢁAndrea, L. Miarelli, Carbohydr.
Res. 1995, 274, 197 ± 208.
[30] H.-M. Liu, Y. Sato, Y. Tsuda, Chem. Pharm. Bull. 1993, 41, 491 ± 501.
[31] S. J. Angyal, V. A. Pickles, Aust. J. Chem. 1972, 25, 1711 ± 1718.
[32] K. Bock, M. Beck Sommer, Acta Chem. Scand. Ser. B 1980, 34, 389.
[33] a) E. Fischer, Ber. Dtsch. Chem. Ges. 1889, 22, 87 ± 97; b) P. A.
Garcia Ruiz, I. Cartagena Travesedo, A. Soler Martinez, Can. J.
Chem. 1984, 62, 870 ± 877
[34] A. A. H. Al-Jobore, R. D. Guthrie, R. D. Wells, Carbohydr. Res. 1971,
6, 474 ± 476; K. Izumori, T. Miyoshi, S. Tokuda, K. Yamabe, Appl.
Environ. Microbiol. 1984, 48, 1055 ± 1057; T. Szumilo, FEBS Lett.
1981, 124, 270 ± 272; O. Eyrisch, G. Sinerius, W. D. Fessner, Carbohydr.
Res. 1993, 238, 287 ± 306.
[53] M. Harenbrock, A. Matzeit, H. J. Schäfer, Liebigs Ann. Chem. 1996,
55 ± 62.
[54] P. Angibeaud, C. Bosso, J. Defaye, D. Horton, C. Cohen-Addad, M.
Thomas, J. Chem. Soc. Perkin Trans. 1 1979, 1583 ± 1592.
[55] L. Slechta, L. E. Johnson, J. Antibiot. 1976, 29, 685 ± 687; L.
Baczynskyj, H. Hoeksema, J. Antibiot. 1976, 29, 688 ± 691; I. A.
Kennedy, T. Remscheidt, J. F. Britten, I. D. Spenser, Can. J. Chem.
1995, 73, 1329 ± 1337.
[56] R. McDaniel, C. R. Hutchinson, C. Khosla, J. Am Chem. Soc. 1995,
117, 6805 ± 6810.
[35] T. Vuorinen, A. S. Serianni, Carbohydr. Res. 1990, 207, 185 ± 210.
[36] For the l enantiomer see ref. [31b].
[57] a) A. Huwig, PhD thesis, University of Saarbrücken 1997; b) M.
Rudrum, D. F. Shaw, J. Chem. Soc. 1965, 52 ± 57; R. U. Lemieux, J. D.
Stevens, Can. J. Chem. 1966, 44, 249 ± 262; c) D. E. Dorman, J. D.
Roberts, J. Am. Chem. Soc. 1970, 92, 1355 ± 1361; d) A. S. Perlin, B.
Casu, H. J. Koch, Can. J. Chem. 1970, 48, 2596 ± 2666; e) T. Vuorinnen,
A. S. Serianni, Carbohydr. Res. 1990, 209, 13 ± 31.
[58] H. Weenen, S. B. Tjan, Dev. Food. Sci. 1994, 35, 327 ± 337.
[60] C.-H. Wong, F. P. Mazenod, G. M. Whitesides, J. Org. Chem. 1983, 48,
3493 ± 3497; K.-C. Liu, T. Kajimoto, L. Chen, Z. Zhong, Y. Ichikawa,
C.-H. Wong, J. Org. Chem. 1991, 56, 6280 ± 6289.
[61] G. Massiot, C. Lavaud, V. Besson, Bull. Soc. Chim. Fr. 1990, 127, 440 ±
445.
[62] I. I. Cubero, M. T. P. Lopez-Espinosa, Carbohydr. Res. 1986, 154, 71 ±
80.
[63] A. D. Backstrom, R. A. S. McCordie, T. P. Begley, J. Carbohydr.
Chem. 1995, 14, 171 ± 175.
[64] C. Bernasconi, L. Cottier, G. Descotes, G. Remy, Bull. Soc. Chim. Fr.
1979, Pt. 2, 5 ± 6, 332 ± 336.
[37] a) G. Machell, G. N. Richards, J. Chem. Soc. 1960, 1938 ± 1944; H.
Kato, Agric. Biol. Chem. 1960, 26, 187 ± 192; b) E. F. L. J. Anet, J. Am.
Chem. Soc. 1960, 82, 1501; E. F. L. J. Anet, Aust. J. Chem. 1960, 132,
396 ± 403; c) M. A. Madson, M. S. Feather, Carbohydr. Res. 1981, 94,
183 ± 191.
[38] M. J. Robins, J. S. Wilson, J. Am. Chem. Soc. 1981, 103, 932 ± 933.
[39] 13C NMR of 3-deoxy-d-fructose: H. Kato, N. Van Chuyen, T. Shinoda,
F. Sekiya, F. Hayase, Biochem, Biophys. Acta 1990, 1035, 71 ± 76;
1H NMR: ref. [40]. 13C NMR of 3-deoxy-d-glucose: P. E. Pfeffer, F. W.
Parrish, J. Unruh, Carbohydr. Res. 1980, 84, 13 ± 23; 1H NMR: ref. [29]
and [40]. 13C NMR of 3-deoxy-d-mannose: C. du Mortier, R. M.
de Lederkremer, J. Carbohydr. Chem. 1984, 3, 219 ± 228.
[40] S. Freimund, S. Köpper, Carbohydr. Res. 1998, 308, 195 ± 200.
[41] S. M. Andersen, I. Lundt, J. Marcussen, S. Yu, I. Sotofte, J. Carbohydr.
Chem. 1998, 17, 1027 ± 1035.
[42] a) A. Baute, R. Baute, G. Deffieux, Phytochemistry 1988, 27, 3401 ±
3403; b) T. Taguchi, M. Haruna, J. Okuda, Biotechnol. Appl. Biochem.
[65] Y. Tsuda, M. Hanajima, N. Matsuhira, Y. Okuno, K. Kanemitsu,
Chem. Pharm. Bull. 1989, 37, 2344 ± 2350.
Â
1993, 18, 275 ± 283; c) S. Yu, L. Kenne, M. Pedersen, Biochim.
Biophys. Acta 1995, 1244, 1 ± 9.
[66] J. S. Brimacombe, A. Husain, Carbohyd. Res. 1968, 6, 491 ± 493; K.
Bock, C. Pedersen, Adv. Carbohydr. Chem. Biochem. 1983, 41, 27 ± 66.
[67] V. S. Rao, J. F. Foster, J. Phys. Chem. 1965, 69, 636 ± 640.
[68] S. Jacobsen, O. Mols, Acta Chem. Scand. Ser. B. 1981, 35, 163 ± 168.
[69] S. A. Barker, J. Homer, M. C. Keith, L. F. Thomas, J. Chem. Soc. 1963,
1538 ± 1543; E. G. Gros, I. O. Mastronardi, A. R. Frasca, Carbohydr.
Res. 1970, 16, 232 ± 234; D. Welti, J. Chem. Res. (M) 1977, 3566 ± 3587.
[43] P. Jarglis, F. W. Lichtenthaler, Angew. Chem. 1982, 94, 140 ± 141;
Angew. Chem. Int. Ed. Engl. 1982, 21, 141 ± 142.
[44] R. K. Ness, H. G. Fletcher, Jr., C. S. Hudson, J. Am. Chem. Soc. 1950,
72, 4547 ± 4549.
[45] D. Y. Gagnaire, F. R. Taravel, M. R. Vignon, Carbohydr. Res. 1976, 51,
157 ± 168.
[46] H. J. Dauben, W. L. Evans, J. Am. Chem. Soc. 1938, 60, 886 ± 890.
Chem. Eur. J. 1998, 4, No. 12
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0947-6539/98/0412-2455 $ 17.50+.50/0
2455