Oligosaccharide synthesis by dextransucrase: New unconventional acceptors

Article abstract of DOI:10.1016/S0008-6215(02)00272-0

The acceptor reactions of dextransucrase offer the potential for a targeted synthesis of a wide range of di-, tri- and higher oligosaccharides by the transfer of a glucosyl group from sucrose to the acceptor. We here report on results which show that the synthetic potential of this enzyme is not restricted to 'normal' saccharides. Additionally functionalized saccharides, such as alditols, aldosuloses, sugar acids, alkyl saccharides, and glycals, and rather unconventional saccharides, such as fructose dianhydride, may also act as acceptors. Some of these acceptors even turned out to be relatively efficient: α-D-glucopyranosyl-(1→5)-D-arabinonic acid, α-D-glucopyranosyl-(1→4)-D-glucitol, α-D-glucopyranosyl-(1→6)-D-glucitol, α-D-glucopyranosyl-(1→6)-D-mannitol, α-D-fructofuranosyl-β-D-fructofuranosyl-(1,2′:2,3′)- dianhydride, 1,5-anhydro-2-deoxy-D-arabino-hex-1-enitol ('-glucal'), and may therefore be of interest for future applications of the dextransucrase acceptor reaction.

Full text of DOI:10.1016/S0008-6215(02)00272-0

Carbohydrate Research 337 (2002) 1811–1820
www.elsevier.com/locate/carres
Oligosaccharide synthesis by dextransucrase: new unconventional
acceptors
Kristin Demuth, Hans-Joachim Jo¨rdening, Klaus Buchholz*
Chair for Carbohydrate Technology, Technical Uni6ersity, Langer Kamp 5, D-38106 Braunschweig, Germany
Received 20 September 2001; received in revised form 28 July 2002; accepted 4 August 2002
Dedicated to Professor Dr. Dr. h.c. Frieder W. Lichtenthaler on the occasion of his 70th birthday
Abstract
The acceptor reactions of dextransucrase offer the potential for a targeted synthesis of a wide range of di-, tri- and higher
oligosaccharides by the transfer of a glucosyl group from sucrose to the acceptor. We here report on results which show that the
synthetic potential of this enzyme is not restricted to ‘normal’ saccharides. Additionally functionalized saccharides, such as
alditols, aldosuloses, sugar acids, alkyl saccharides, and glycals, and rather unconventional saccharides, such as fructose
dianhydride, may also act as acceptors. Some of these acceptors even turned out to be relatively efficient: a-
5)- -arabinonic acid, a- -glucopyranosyl-(1“4)- -glucitol, a- -glucopyranosyl-(1“6)- -glucitol, a- -glucopyranosyl-(1“6)-
-mannitol, a- -fructofuranosyl-b- -fructofuranosyl-(1,2%:2,3%)-dianhydride, 1,5-anhydro-2-deoxy- -arabino-hex-1-enitol (‘ -glu-
D-glucopyranosyl-(1“
D
D
D
D
D
D
D
D
D
D
D
cal’), and may therefore be of interest for future applications of the dextransucrase acceptor reaction. © 2002 Elsevier Science Ltd.
All rights reserved.
Keywords: Oligosaccharide synthesis; Glycosylation; Dextransucrase; Acceptor reactions
1. Introduction
lently bound glucosyl residue, is released from the
active site by acceptor reactions.9,10 Nevertheless, at the
molecular level, the catalytic action of dextransucrase is
still not understood in detail, since all attempts to
crystallize this enzyme have failed thus far. However
the amino acid sequences of some dextransucrases have
recently been reported and the probable location of the
catalytic sites identified by sequence comparison and
site-directed mutagenesis.11,12
Dextransucrase (EC 2.4.1.5) is a bacterial enzyme
produced extracellularly by various bacterial species of
Leuconostoc, Lactobacillus, and Streptococcus which
catalyzes two kinds of reaction: (1) the primary reaction
(substrate reaction), the synthesis of dextran from su-
crose; and (2) the secondary reaction (acceptor reac-
tion), the transfer of
D-glucose from sucrose to
carbohydrate acceptors that are added. The mecha-
nisms for these reactions were studied first by Ebert and
co-workers and later extensively by Robyt and co-
workers.1–8 They proposed that dextran is synthesized
The acceptor reaction has been of special interest to
us in that it has many possible technical applications
for the synthesis of new oligosaccharide derivatives.
Many different mono-,2,10,13–17 di-,10,15,18–21 tri-,5,10,13,22
and oligo-23,24 saccharides may act as acceptors for
dextransucrase. With the exception of a few acceptors,
the acceptor product itself can serve as an acceptor,
leading to the formation of an homologous series of
isomalto-oligosaccharides attached to the acceptor,
which is located at the reducing end of the product. The
efficiency and extent of the acceptor reaction varies,
depending on the particular acceptor.10,17,20,23–25 So far,
by a two-site insertion mechanism in which the
D-glu-
cosyl group is added to the reducing-end of a covalently
linked, growing dextran chain. The chain, or the cova-
* Corresponding author. Tel.: +49-531-380-0920; fax: +
49-531-380-0988
E-mail address: k.buchholz@tu-bs.de (K. Buchholz).
0008-6215/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0008-6215(02)00272-0

1812
K. Demuth et al. / Carbohydrate Research 337 (2002) 1811–1820
alditols,
charide derivatives a-
tol (3, ‘maltitol’), a- -glucopyranosyl-(1“6)-
(4), and a- -glucopyranosyl-(1“6)-
-mannitol (5);‡
D
-glucitol (1)† and
-glucopyranosyl-(1“4)-
-glucitol
D
-mannitol (2), the disac-
the glucodisaccharides maltose and isomaltose were
found to be the best acceptors, whereas several other
saccharides do not seem to be acceptors at
D
D-gluci-
D
D
all.10,13,15,17,20,21,26,27 For many acceptors,
D-glucose is
D
D
transfered to the 6-hydroxyl group to give an a-(1“6)
glycosidic linkage, but all of the other possible a-glyco-
sidic linkages also have been observed to be formed,
e.g., (1“1),27–29 (1“2),10,15,30–35 (1“3),17,35 (1“4),17
and (1“5),22,26 depending on the particular structure
of the acceptor.
To date, there has not been a detailed explanation as
to why there are these differences in bond formation
specifity as well as in acceptor efficiency, and only few
systematic investigations concerning the relationships
between the molecular structure of a given saccharide
and its acceptor properties have been carried out
(methyl a-D
-glucopyranoside analogues,17 maltooligo-
saccharides23,24 and gluco-disaccharides20). Moreover,
little is known about the acceptor properties of deriva-
tized saccharides, although numerous saccharide
derivatives have been tested as inhibitors for dextransu-
crase, for example, fluoro saccharides,36–42 chloro sac-
charides,43,44 amino saccharides,45–47 and thio
saccharides.39,40,42,47 Our interest was in studying the
potential of other saccharides and their derivatives as
acceptors to give new products, and to demonstrate
further synthetic potential of the dextransucrase gluco-
sylation reaction. The focus here is on such saccharides
as alditols, aldosuloses, aldonic acids, fructose dianhy-
drides, and glycals. They may be building blocks for
further synthesis carrying the potential for enhanced
solubility, hydrophilicity, and reduced toxicity.
On the basis of our kinetic model for dextransu-
crase,27,48–50 a kinetic definition for the strength of the
acceptor was developed. This method is in contrast to
the method based on the final yield of product used by
Robyt et al.10,17 It has the advantage of being indepen-
dent of both the initial saccharide concentration and of
the eventual acceptor efficiency of the products formed.
The strength of the acceptors was determined by carry-
ing out at least one kinetic experiment, using iterative
computation of the kinetic parameters.
the aldosuloses derived from sucrose and its (regio-)
isomers, by microbial oxidation of the 3-position of the
glucopyranosyl group, a-
osyl-b- -fructofuranoside (6, ‘3-keto-sucrose’), a-
bino-hexos-3-ulopyranosyl-(1“5)- -fructose (7), a-
arabino-hexos-3-ulopyranosyl-(1“6)- -fructose (8),
and the oxidation product of 5, a- -arabino-hexos-3-
ulopyranosyl-(1“6)- -mannitol (9); the sugar acid,
D-arabino-hexos-3-ulopyran-
D
D-ara-
D
D-
D
D
2. Results and discussion
D
In the dextransucrase acceptor reaction, sucrose was
used as the substrate, where the glucosyl group is
transfered, by tranglycosylation, to the acceptor and
fructose is released as a product (see graphical ab-
stract). Dextran is formed in major or minor amounts,
depending on the strength of the acceptor and the
reaction conditions. The following acceptors were
tested:
† Earlier, Ebert and co-workers61,62 reported from their
observation, that the two hexitols glucitol and mannitol cause
a decrease in the average molecular weight of the dextran
formed during dextransucrase catalysis. They therefore as-
sumed both these hexitols to be dextransucrase acceptors,
although they never succeeded in detecting the corresponding
low molecular weight acceptor products.
‡ Evidence for acceptor properties of this saccharide was
first obtained by D. Prinz.21

K. Demuth et al. / Carbohydrate Research 337 (2002) 1811–1820
1813
the fructose dianhydride a-
D
-fructofuranosyl-b-D-
fructofuranosyl-(1,2%:2,3%)-dianhydride (16); and
the sucrose regioisomers a-
-fructose (‘maltulose’, 17), a-
6)- -fructose (‘isomaltulose’, 18), and a-
pyranosyl-(1“1)- -fructose (‘trehalulose’, 19).
D
-glucopyranosyl-(1“4)-
-glucopyranosyl-(1“
-gluco-
D
D
D
D
D
the oxidation product from a-
D
-glucopyranosyl-(1“6)-
-glucopyranosyl-(1“5)-
D
-fructose (‘isomaltulose’), a-
D
D-arabinonic acid (10, evidence for acceptor properties
of this saccharide was first obtained by D. Prinz21);
Primary results from the tests of acceptor proper-
ties.—The chromatograms obtained by analysis with
either the HPAEC system (Fig. 1) or the aminopropyl
silica gel column and the cation exchanger column in
the silver form clearly shows that all alditols, two
aldosuloses, 8 and 9, a-
binonic acid (10), one glycal, 11, the fructose dianhy-
dride 16, and the sucrose analogue a- -gluco-
pyranosyl-(1“4)- -fructose (17), gave rise to a ho-
mologous series of acceptor products. Furthermore, it
was especially remarkable that we were able to detect
two different primary acceptor products both for the
D
-glucopyranosyl-(1“5)-
D-ara-
the glycals 1,5-anhydro-2-deoxy-
tol (‘ -glucal’, 11), ethyl 2,3-dideoxy-a-
enopyranoside (‘ethyl pseudoglucal’, 12), 2,6-anhydro-
1-deoxy- -arabino-hex-1-enitol (13);
D
-arabino-hex-1-eni-
D
D
D-erythro-hex-2-
D
D
alditol,
D-glucitol, and the sucrose regioisomer 17. The
two other sucrose regioisomers (18 and 19) were al-
ready known to act as acceptors.21 Remarkably, in case
of the acceptor reaction of 17, one of the two primary
products was completely consumed during the course of
the reaction.
For the two glycals, 12 and 13, and the two remain-
ing aldosuloses, 6 and 7, we did not succeed in detect-
ing any acceptor product using any of our
chromatographic systems. It must be assumed, there-
fore, that these saccharide derivatives are not acceptors
in the dextransucrase reaction. In the case of the two
glycals, the lack of verifiable acceptor reactions may be
explained by conformational changes in the pyranoside
rings, caused by the presence of a double bond. More-
over, the hydroxyl groups, which are missing in these
molecules, usually play a key role in the acceptor
reaction of gluco- and fructo-pyranosyl saccharides. In
a systematic investigation, Robyt et al.17 have shown
that the 2-, 3- and 4-hydroxyl groups play an important
role for the acceptor properties of glucopyranose ana-
logues. In the case of the two aldosuloses, we indeed
did not expect to observe any acceptor products, as the
the n-alkyl glucosides n-octyl b-
D
-glucopyranoside
(14) and n-decyl [a-
D
-glucopyranosyl-(1“4)-] a-D-
glucopyranoside (15);
unmodified disaccharides, sucrose and its isomer, a-D-

1814
K. Demuth et al. / Carbohydrate Research 337 (2002) 1811–1820
Fig. 1. HPAEC chromatogram from a sample of an acceptor reaction digest of a-D-glucopyranosyl-(1“5)-D-arabinonic acid (10).
glucopyranosyl-(1“5)-
act as acceptors for dextransucrase.
For n-alkyl glycosides, we were not able to discover
D
-fructose (‘leucrose’), do not
on the other hand, does not have such symmetry and
two different products should result, a- -glucopyran-
osyl-(1“6)- -glucitol (4), from reaction of the 6-hy-
droxyl group and a- -glucopranosyl-(1“6)- -gulitol
from that of the 1-hydroxyl group. A comparison of the
retention times on HPAEC under various analytical
conditions showed that the primary acceptor product of
mannitol was indeed 5, as well as the higher homo-
logues of the mannitol acceptor product series, which
showed the same retention times as those of the accep-
tor-product series of 5. Likewise, one of the two pri-
D
D
any new products in the case of n-octyl b-
ranoside (14), whereas the chromatograms from the
reactions with n-decyl [a- -glucopyranosyl-(1“4)-] a-
-glucopyranoside (15) revealed the formation of a
D-glucopy-
D
D
D
D
small amount of a new product: A new signal appeared
in the region of the chromatograms where the corre-
sponding primary acceptor product was expected.
Structure of the primary products from
-mannitol.—As already mentioned, there were two
different primary acceptor products detected in case of
-glucitol, whereas mannitol only gave single
D-glucitol and
D
mary acceptor products of D-glucitol was identified as
4. On basis of these findings, it therefore stands to
reason that the hexitols are able to react as acceptors
with both their 1- and 6-hydroxyl groups. It must be
pointed out however that the acceptor reaction of the
hexitols is much slower, by one order of magnitude,
than that of the first (and subsequent two) acceptor
products, notably the disaccharide alcohols 4 and 5 (see
D
a
product. Assuming that both of these alditols are capa-
ble of reacting by nucleophilic attack of either the 1- or
6-hydroxyl group, this observation may be explained as
follows: Due to the symmetry of the mannitol molecule
only a single primary product is possible, a-
D-glucopy-
ranosyl-(1“6)- -mannitol (5). The -glucitol molecule,
D
D
later, Fig. 3). No di-D-glucopyranosyl product was

K. Demuth et al. / Carbohydrate Research 337 (2002) 1811–1820
1815
found. This is reasonable, since the first acceptor prod-
ucts—the -glucopyranosyl-glucitol, or -mannitol, re-
spectively, are further glucosylated at 6-OH; this
reaction is much more rapid, as compared to the gluco-
sylation of the sugar alcohols 1 and 2.
NMR characterization of selected acceptor prod-
ucts.—In order to gain more information about the
molecular structure of the new acceptor products, the
proton of the a-(1“6)-glycosidic bond between the
fructofuranosyl and the modified glucopyranosyl
residue and the expected 13C NMR signals at 98.8, 98.6,
and 66.4 ppm (two peaks) confirmed the existence of
two a-(1“6)-glucosidic linkages between the glucosyl
residues. By analogy, the secondary product 16b gave,
when compared to the spectrum of of the fructose
dianhydride 16 itself, two additional characteristic dou-
D
1
primary acceptor products of a-
5)- -arabinonic acid (10) and 1,5-anhydro-2-deoxy-
arabino-hex-1-enitol (‘ -glucal’, 11), as well as the
D-glucopyranosyl-(1“
blets in the H NMR spectrum. These doublets were
D
D-
centered at 4.99 and 5.01 ppm, both showing a coupling
3
constant of J 3.7 Hz. The corresponding 13C NMR
D
secondary acceptor products of the aldosulose 8 (the
isolated secondary acceptor product of 8 still contained
some primary acceptor product as contamination) and
the fructose dianhydride 16 were isolated from the
reaction mixtures by preparative chromatography and
signals occured at 99.2, 98.6 ppm (C-1 of the two
transferred glucopyranosyl residues) and 67.6 and 66.3
ppm, indicating that C-6 of the a-fructofuranosyl moi-
ety and C-6 of the first transfered glucopyranosyl
residue act as nucleophiles; comparison with the NMR
data of 16 itself revealed that the molecule must have
reacted by nucleophilic attack of the 6-hydroxyl group
of the a-fructofuranosyl moiety.
1
then analyzed by H and 13C NMR spectroscopy.
1
The resulting H NMR spectra of the primary prod-
ucts 10a and 11a from 10 and 11 both showed, when
compared to the original spectra of 10 and 11, a new
doublet at l 5.01 and 5.00 ppm with a coupling con-
stant of 3.7 Hz, which according to Bradbury et al.55 is
characteristic for the anomeric proton of an a-(1“6)-
glucosidic linkage. Moreover, the 13C NMR spectra
also displayed the corresponding characteristic sig-
nals.35,56–58 In the case of the 11, the signal for C-6 of
the glucosyl moiety carrying the double bond appeared
at 66.4 ppm and the signal of the C-1% of the transfered
glucosyl residue at 99.0 ppm. For the acceptor product
of 10, analogous signals for C-6% and C-1¦ were found
at 66.5 and 98.7 ppm. It can thus be concluded that
both 10 and 11 react as ‘normal’ acceptors, namely by
nucleophilic attack of their 6-hydroxyl group.
Further experimental e6idence for the molecular struc-
ture of the acceptor products.—Additional evidence
concerning the structure of the various acceptor prod-
ucts was obtained by HPAEC chomatography: Samples
from dextransucrase acceptor reaction digests of glu-
cose, maltose and a-
tose (‘isomaltulose’, 18) and
a- -glucopyranosyl-(1“6)-
[as well as a- -glucopyranosyl-(1“6)-
a- -glucopyranosyl-(1“4)- -glucitol (3), a-
ranosyl-(1“5)- -arabinonic acid (10), 11 (‘
the fructose dianhydride 16, the sucrose analogues a-
glucopyranosyl-(1“4)- -fructose (‘maltulose’, 17) and
D
-glucopyranosyl-(1“6)-
-glucitol (1) [as well as
-glucitol, 4] -mannitol (2)
-mannitol, 5],
-glucopy-
-glucal’),
D-fruc-
D
D
D
D
Similar results were obtained from the NMR investi-
D
D
gations of the secondary acceptor products 8b and 16b
1
D
D
D
obtained from 8 and 16. In the H NMR spectrum of
D
D
the acceptor product 8b, the presence of two more
doublets centered at 5.01 and 5.07 ppm in addition to
the doublets at 5.41 ppm resulting from the anomeric
D-
D

1816
K. Demuth et al. / Carbohydrate Research 337 (2002) 1811–1820
Fig. 2. Logarithmic capacity factors obtained from the HPAEC system under isocratic elution conditions (98% of eluent A and
2% of eluent B) for various dextransucrase acceptors and their respective acceptor products as a function of the degree of
polymerization; correlation of numbers with saccharides:
(‘maltitol’, 3), a- -glucopyranosyl-(1“6)- -glucitol (4), a-
arabinonic acid (10), 1,5-anhydro-2-deoxy- -arabino-hex-1-enitol (‘
(1,2%:2,3%)-dianhydride (16), a- -glucopyranosyl-(1“4)- -fructose (‘maltulose’, 17), a-
(isomaltulose, 18), a- -glucopyranosyl-(1“1)- -fructose (‘trehalulose’, 19).
D
-glucitol (1),
-glucopyranosyl-(1“6)-
-glucal’, 11), a-
D
-mannitol (2), a-
-mannitol (5), a-
-fructofuranosyl-b-
-glucopyranosyl-(1“6)-D
D
-glucopyranosyl-(1“4)-
-glucopyranosyl-(1“5)-
-fructofuranosyl-
-fructose
D-glucitol
D
D
D
D
D
D-
D
D
D
D
D
D
D
D
D
a-
D
-glucopyranosyl-(1“1)-
D
-fructose
(‘trehalulose’,
analogue a-
D
-glucopyranosyl-(1“4)-D-fructose (‘mal-
19) were analyzed under specific isocratic elution condi-
tions (98% of eluent A and 2% of eluent B). Then the
logarithmic capacity factors§ of the various acceptors
and their respective acceptor products were calculated
from the observed retention times and plotted versus
the degree of polymerization. As may be seen from Fig.
2, straight lines were obtained for each of the ho-
mologous series of acceptor products and—what is
especially remarkable—the slopes of all these lines were
identical (see also Table 1).
tulose’, 17) acceptor reaction digest showed that 17
gives rise to two different primary acceptor products.
According to our findings from the logaritmic plot of
capacity factors (preceding paragraph), one of these
two acceptor products should be a ‘normal’ one, that is
17 should have reacted by nucleophilic attack of the
6-hydroxyl group of its glucopyranosyl residue. The
second primary acceptor product initially is formed
during the course of the reaction and later disappears
completely from the reaction mixture. It may be as-
sumed that 17 reacts also as an acceptor in its b-fructo-
It is well known that glucose, maltose and 18 (‘iso-
maltulose’) all give a series of ‘normal’ acceptor prod-
ucts with isomalto oligosaccharide groups attached via
a glycosidic bonds to the 6-hydroxyl group of the
non-reducing glucosyl moiety.5,10,15 This observation
suggests that the observed slope is characteristic for a
series of oligosaccharides with a-(1“6)-linked glucopy-
ranosyl residues under the given analytical conditions.
We also found a completely different elution behavior
in the case of the maltooligosaccharide series, where all
Table 1
Results from the logarithmic plot of the acceptor products
capacity factors versus their degree of polymerization
Acceptor
Slopea Intercept
D
D
D
-Glucose (and isomaltose)
-Glucitol 1 (and 4)
-Mannitol 2 (and 5)
0.575
0.591
0.579
0.417
0.574
0.569
0.574
0.437
−0.907
−0.730
0.589
0.396
0.927
D-glucopyranosyl residues are a-(1“4)-linked. In other
3 (‘Maltitol’)
17 (‘Maltulose’)
16 (Fructose dianhydride)
words, these findings can be interpreted as another
indication that all acceptors, which were taken into
consideration in this investigation, react in the ‘normal’
way.
a-
D
-Glucopyranosyl-(1“5)-
-arabinonic acid (10)
1.677
D
As mentioned in the first paragraph of this paper, the
11 (
D
-‘Glucal’)
0.579
0.574
0.565
0.586
−1.441
1.565
0.936
HPAEC chromatograms obtained from the sucrose
Maltose
18 (‘Isomaltulose’)
19 (‘Trehalulose’)
§ The capacity factor ki for any substance i was determined
according to the equation ki=(tr,i−t0)/t0 (tr,i: retention time
of substance i under the given analytical conditions; t0: dead
time of the analytical system).
0.570
a Average slope of lines: m=0.576 (90.002) (standard
error: 0.4%).

K. Demuth et al. / Carbohydrate Research 337 (2002) 1811–1820
1817
Fig. 3. Acceptor strength of various mono- and di-saccharides and their acceptor products in relation to the acceptor efficiency
of
a-
D
-fructose (logarithmic scale; DP: degree of polymerization); for correlation of numbers with saccharides see Fig. 2; further:
-arabino-hexos-3-ulopyranosyl-(1“6)- -fructose (8), a- -arabino-hexos-3-ulopyranosyl-(1“6)- -mannitol (9), n-decyl [a-
-glucopyranoside (15).
D
D
D
D
D-
glucopyranosyl-(1“4)-] a-
D
furanoside form via the 2-hydroxyl group of the fructo-
syl moiety to give a- -glucopyranosyl-(1“4)-b- -fruc-
tofuranosyl-(2“1)-a- -glucopyranoside as the second
primary acceptor-product. This follows from the struc-
tural analogy to b- -galactopyranosyl-(1“4)- -fruc-
tose (‘lactulose’, 20) which is known to give an acceptor
product bound similarly,33 which in turn is a substrate
for dextransucrase, giving dextran and 20 as reaction
products.34 This would explain the consumption of one
of the acceptor products of 17 during the course of the
reaction.
Determination of the kinetic parameters and acceptor
efficiencies.—For a given acceptor, the value of the
kinetic parameter describing the rate of the acceptor
reaction step itself can be viewed as a general criterion
for its acceptor strength. In this work, we decided to
determine the kinetic parameters for the acceptor reac-
tion of not only each of the new acceptors themselves
(namely the various saccharide derivatives, which were
initially added to the reaction mixtures), but also of
their acceptor products. Therefore, the kinetic model
for the dextransucrase catalysis, which we have pre-
sented in several earlier publications,27,49,50 had to be
extended: new parameters were introduced for the ac-
ceptor reactions of all members of the homologous
series of acceptor products. Moreover, potential inhibi-
tion caused by the acceptor products were also taken
into consideration in the extended model.
sidered) concentrations into account. (For a more de-
tailed descripition of the calculation prodedure see
Refs. 50 and 59.) Fig. 3 presents the main results
obtained from these calculations, namely the acceptor
reaction parameters estimated for each of the new
acceptors and also those of the respective acceptor
products in relation to that of fructose. Moreover, for
better comparison, Fig. 3 also shows our previous
modeling results for the glucose and the maltose
acceptor.50,60
D
D
D
D
D
As may be seen from Fig. 3, maltitol (3) is the best
acceptor among the saccharide derivatatives studied in
our investigations. Nevertheless, its acceptor efficiency
is considerably less than that of a-
(1“6)- -glucose (22, ‘isomaltose’) and maltose (23),
which were the strongest acceptors known so far.10,23–25
Compound 3 is followed by the sugar acid a- -glucopy-
ranosyl-(1“5)- -arabinonic acid (10), and then by a-
-glucopyranosyl-(1“6)- -mannitol (5), the fructose
dianhydride 16, a- -glucopyranosyl-(1“6)- -glucitol
(4) and 11 (‘ -glucal’). The remaining new acceptors,
the aldosuloses 8 and 9, mannitol, -glucitol and the
D-glucopyranosyl-
D
D
D
D
D
D
D
D
D
n-alkyl glycoside 15 must be considered as weak accep-
tors, since their acceptor strength is below that of
D-fructose.
In accordance with the results obtained by J. F.
Robyt et al.,17 the data given in Fig. 3 generally illus-
trate, that the unchanged glucopyranosyl group at the
non-reducing-end of a saccharide is the best structural
component suited for dextransucrase acceptors. Among
The kinetic parameters for the acceptor reaction for
each of the acceptors were then determined by iterative
parameter estimation, taking all data for sucrose, fruc-
all monosaccharides examined here,
D-glucose was
tose, a-
D
-glucopyranosyl-(1“5)-
D
-fructose (21, ‘leu-
found to be the best acceptor, followed by 11 (‘
cal’), which is the only other monosaccharide here with
D
-glu-
crose’), acceptor and acceptor product (only the first
four members of the acceptor product series were con-
an analogous glucopyranosyl molecular structure.

1818
K. Demuth et al. / Carbohydrate Research 337 (2002) 1811–1820
Moreover, the same tendency was also observed in case
of the disaccharides. The acceptor efficiency of the
disaccharides 20 (‘isomaltose’) and maltose 23, which
both have two unchanged glucopyranosyl units, was
considerably higher than that of the disaccharides hav-
ing only one unmodified glucopyranosyl unit, namely 3
(‘maltitol’), 10, 17 (‘maltulose’), and the disaccharide
alditols 4 and 5. The acceptor strength of the two
aldosuloses derived from disaccharides, 8 and 9 with
both glucopyranosyl units modified were found to be
only somewhere in the order of that of the
monosaccharides.
glucopyranosyl-(1“5)- -fructose (‘leucrose’, 21), how-
D
ever, the fructosyl group is fixed strictly in the a- or
b-pyranoside form, and in the sucrose molecule to the
b-furanoside form.
Summarizing our investigations, the acceptor reac-
tions of dextransucrase offer the potential for a targeted
synthesis of a wide range of di-, tri- and higher
oligosaccharides by the transfer of glucose to the accep-
tor. The synthetic potential of this enzyme is not re-
stricted to ‘normal’ saccharides: Also additionally
functionalized saccharides, such as alditols (1–4), aldo-
suloses (8, 9), a sugar acid (10),
D-glucal (11), and
On the whole, the most striking evidence for the
good suitability of the glucopyranosyl (and the isomal-
tosyl) unit was obtained from our findings that the
subsequent higher homologue acceptor products—es-
pecially the trisaccharides—are altogether very good
acceptors for dextransucrase. These modeling results do
not mean, however, that all saccharides containing an
unaffected glucopyranosyl residue necessarily have to
be relatively good acceptors for dextransucrase. In-
stead, it must be kept in mind, that every attached
group may have a severe influence on the given saccha-
ride’s acceptor properties. In this work, for instance,
the presence of long n-alkyl chains turned out to be
definitely disadvantegous for the acceptor properties of
saccharide derivatatives.
rather unconventional saccharides, such as the fructose
dianhydride 16, may act as acceptors. Some of these
acceptors even turned out to be relatively efficient (e.g.,
10, 3–5, 16, 11) and may therefore be of interest for
future applications of the dextransucrase acceptor
reaction.
3. Experimental
Enzyme.—All experimental investigations were car-
ried out with the dextransucrase from Leuconostoc
mesenteroides NRRL B-512F, since the enzyme from
this species shows a high specifity in bond formation.
The enzyme was a gift from the Pfeiffer & Langen
company, Cologne, Gemany. For determination of the
dextransucrase activity, the maltose test (see below)51
was used.
Another example, demonstrating the importance of
attached groups and their structural arrangement is
provided from the series of sucrose analogues. Among
these, both a-
D
-glucopyranosyl-(1“6)-
D
-fructose (‘iso-
Substrates and acceptors.—The sugar alcohols a-
glucopyranosyl-(1“6)- -glucitol (4) and a- -glucopy-
ranosyl-(1“6)- -mannitol (5) were a gift from the
‘Su¨dzucker AG’, Mannheim, Germany, and a-
glucopyranosyl-(1“4)- -fructose (‘maltulose’, 17) was
D-
maltulose’, 18) and a-
D-glucopyranosyl-(1“1)-
D-fruc-
D
D
tose (‘trehalulose’, 19) have, according to the results of
Prinz,21 to be classified as strong acceptors. In contrast,
D
D-
a-
D
-glucopyranosyl-(1“5)-
D-fructose (‘leucrose’) and
D
sucrose itself do not seem to be able to act as acceptors
a gift from Professor F. W. Lichtenthaler (Technical
University, Darmstadt, Germany).
at all.10 The two remaining members of the sucrose
analogue series, a-
D
-glucopyranosyl-(1“3)-
D
-fructose
All of the aldusuloses (6–9) were synthesized by
microbial oxidation with Agrobacterium tumefaciens.52
(‘turanose’)10 and a-
D-glucopyranosyl-(1“4)-
D-fruc-
tose (‘maltulose’, 17), finally, may be looked at as
moderately efficient acceptors. In order to explain the
differences observed, a closer look was taken at the
molecular structure of the various sucrose analogs.
Thereby, a correlation between the acceptor efficiency
and the flexibility of the attached fructosyl residue
concerning structural arrangement became obvious. In
the case of the two strong dextransucrase acceptors, 18
and 19, the fructosyl group is bound very flexibly via a
hydroxymethylene group. The remaining sucrose ana-
logues do not have such a hydroxymethylene-group
linking the two monosaccharide groups. Nevertheless,
a-
D
-Glucopyranosyl-(1“5)-
D
-arabinonic acid (10) and
a-D-fructofuranosyl-b- -fructofuranosyl-(1,2%:2,3%) di-
D
anhydride (16) were gifts from ‘Su¨dzucker AG’ and
‘Nordzucker
AG’
(Braunschweig,
Germany),
respectively.
The two glycals 11 and 12 were both chemically
synthesized from 3,4,6-tri-O-acetyl- -glucal according
to the literature.53 The glycal 13 was prepared from
D
D
-fructose by a four-step synthesis similar to that de-
scribed by Lichtenthaler et al.54 The remaining saccha-
ride derivatatives were commercially available
chemicals.
in the case of 17 (‘maltulose’) and a-
D
-glucopyranosyl-
Experiments for determination of the kinetic parame-
ters and potential acceptor properties.—With all saccha-
rides mentioned, at least one kinetic experiment was
carried out. Standard reaction conditions used were the
following: initial sucrose concentration: 526 mM; initial
(1“3)- -fructose (‘turanose’, 24), the fructosyl group
D
is still somewhat flexible, since in these molecules it may
undergo structural change forming either the a- or
b-furanoside or the a- or b-pyranoside form. In a-D-

K. Demuth et al. / Carbohydrate Research 337 (2002) 1811–1820
1819
concentration of potential acceptor: 920 mM (in the
case of 5 and the alkyl glycosides, 14 and 15, different
initial concentrations had to be chosen because of the
lower water solubility of these saccharides at 25 °C);
dextransucrase activity: 0.80 U/mL; pH 5.4 (maintained
by a 25 mM calcium acetate buffer); 25 °C. 1 Unit is
defined by the formation of 1 mmol of fructose per min,
as measured in the ‘maltose test’ with 0.146 M sucrose
as the substrate and 0.73 M maltose as an acceptor,
other conditions as before.51
mM NaOAc buffer with a pH of 4.7; flow rate: 7.0
mL/min; ambient temperature. Removal of the acetate
from the product fractions was performed by prepara-
tive chromatography on a Sephadex G-10 column
(Pharmacia, Uppsala, Sweden).
NMR spectroscopy.—1H and 13C NMR spectroscopy
was performed with a Bruker AM-400 spectrometer at
400 and 100 MHz, respectively. D2O was used as the
solvent.
The reactions were allowed to proceed for at least 24
h, within this time samples were repeatedly taken from
the reaction mixtures. For determination of the saccha-
ride concentrations and for detection of potential new
acceptor products, these samples were then diluted,
filtered on a membrane, and analyzed by various ap-
propriate HPLC systems.
Acknowledgements
Many discussions and valuable suggestions by Pro-
fessor John Robyt are gratefully acknowledged.
References
HPLC.—Analysis of the samples from assays with
sugar alcohols as well as 17, 10, 16, and glycals were
mainly performed with the highly sensitive HPAEC
system from the Dionex corporation (column: ‘Carbo-
Pak PA 1’, 10 mm, 4×250 cm; eluents: A) 0.1 M
NaOH and B) 1.0 M NaOAc in 0.1 M NaOH; gradient:
in 20 min from 100% A to 97% A, in 5 min from 97%
A to 85% A, 3 min at 85% A; temperature: 30 °C; flow
rate: 1.0 mL/min). Aldusuloses decompose rapidly un-
der strongly basic conditions and thus the HPAEC
system was not suitable for these samples. Therefore,
other HPLC systems were used: an aminopropyl silica
gel column for determination of the mono-, di- and
trisaccharides (column: ‘LiChroSpher 100 NH2’, 5 mm,
4×250 mm, from Merck, Darmstadt, Germany; elu-
ent: 80:20 (v/v) or 70:30 (v/v) acetonitrile–water; flow
rate: 0.8 mL/min; temperature: 22 °C); and a strong
cation exchanger column in the silver form for detec-
tion of higher oligosaccharides (column: ‘Aminex HPX
65 A’, 7.8×300 mm from BioRad Laboratories, Rich-
mond, USA; eluent: distilled water; flow rate: 0.6 mL/
min; temperature: 70 °C). The analytical determination
of the n-alkyl glycosides was performed with an octade-
cyl silica gel column (column: ‘LiChroSpher 100 RP-
1. Ebert, K. H.; Schenk, G. Ad6. Enzymol. 1968, 30, 179–
221.
2. Robyt, J. F. Ad6. Carbohydr. Chem. Biochem. 1995, 51,
133–168.
3. Robyt, J. F.; Kimble, B. K.; Walseth, T. F. Arch.
Biochem. Biophys. 1974, 165, 634–640.
4. Robyt, J. F.; Walseth, T. F. Carbohydr. Res. 1978, 61,
433–445.
5. Robyt, J. F.; Eklund, S. H. Bioorg. Chem. 1982, 11,
115–132.
6. Fu, D.; Robyt, J. F. Carbohydr. Res. 1988, 183, 97–109.
7. Su, D.; Robyt, J. F. Arch. Biochem. Biophys. 1994, 308,
471–476.
8. Robyt, J. F.; Kim, D.; Yu, L. Carbohydr. Res. 1995, 266,
293–299.
9. Robyt, J. F.; Taniguchi, H. Arch. Biochem. Biophys.
1976, 174, 129–137.
10. Robyt, J. F.; Eklund, S. H. Carbohydr. Res. 1983, 121,
279–286.
11. Monchois, V.; Willemot, R.-M.; Remaud-Simeon, M.;
Croux, C.; Monsan, P. Gene 1996, 182, 23–32.
12. Monchois, V.; Remaud-Simeon, Russell, R. R. B.; Mon-
san, P.; Willemot, R.-M. Appl. Microbiol. Biotechnol.
1997, 48, 465–472.
13. Koepsell, H. J.; Tsuchiya, H. M.; Hellmann, N. N.;
Kazenko, A.; Hoffmann, C. A.; Sharpe, E. S.; Jackson,
R. W. J. Biol. Chem. 1953, 200, 793.
14. Jones, R. W.; Jeanes, A.; Stringer, C. S.; Tsuchiya, H. M.
J. Am. Chem. Soc. 1956, 78, 2499–2502.
15. Bailey, R. W.; Barker, S. A.; Bourne, E. J.; Stacey, M. J.
Chem. Soc. 1957, 00, 3530–3536.
16. Bailey, R. W.; Barker, S. A.; Bourne, E. J.; Stacey, M. J.
Chem. Soc. 1957, 00, 3536–3541.
17. Fu, D.; Slodki, M. E.; Robyt, J. F. Arch. Biochem.
Biophys. 1990, 276, 460–465.
18. Killey, M.; Dimler, R. J.; Cluskey, J. E. J. Am. Chem.
Soc. 1955, 77, 3315–3318.
19. Bourne, E. J.; Hutson, D. H.; Weigel, H. Biochem. J.
1961, 79, 549.
18e’,
5 mm, 4×250 mm, from Merck; eluent:
acetonitrile/water in the range 40:60 (v/v) to 65:35 (v/v);
flow rate: 0.8 mL/min, temperature: 22 °C).
Preparati6e chromatography.—For preparative chro-
matography of the reaction mixtures obtained with 10,
11, and
8
the octadecyl silica gel ‘Labomatic
Labochrom Gel RP-18 Silica HD-Sil-18-30-10’ (20–40
mm) (Krannich, Go¨ttingen, Germany) was used as the
stationary phase. Column dimensions were 1–1.3 m×1
cm; eluent: distilled water; flow rate: in the range
0.16–0.27 mL/min; temperature: 2 °C.
The reaction mixture obtained from 10 was separated
by weak anion exchange chromatography on a prepara-
tive aminopropyl silica gel HPLC column (‘LiChroSorb
NH2’, 7 mm, 40×250 mm, from Merck); eluent: 10
20. Yamauchi, F.; Ohwada, Y. Agric. Biol. Chem. 1969, 33,
1295–1300.
21. Prinz, D. Ph.D. Thesis, Technical University of Braun-
schweig, Germany, 1990.
22. Nisizawa, T.; Takeuchi, K.; Imai, S.; Kitahata, S.;
Okada, S. Carbohydr. Res. 1986, 147, 135–144.

1820
K. Demuth et al. / Carbohydrate Research 337 (2002) 1811–1820
23. Fu, D.; Robyt, J. F. Arch. Biochem. Biophys. 1990, 283,
379–387.
24. Fu, D.; Robyt, J. F. Carbohydr. Res. 1991, 217, 201–211.
25. Mayer, R. M.; Matthews, M. M.; Futerman, C. L.;
Parnaik, V. K.; Jung, S. M. Arch. Biochem. Biophys.
1981, 208, 278–287.
44. Tanriseven, A.; Robyt, J. F. Carbohydr. Res. 1989, 186,
87–94.
45. Thaniyavarn, S.; Singh, S.; Maynard, C. M.; Taylor, K.
G.; Doyle, R. J. Carbohydr. Res. 1981, 96, 134–137.
46. Kobayashi, M.; Yokohama, I.; Matsuda, K. Agric. Biol.
Chem. 1986, 50, 2585–2590.
26. Stodola, F. H.; Sharpe, E. S.; Koepsell, H. J. J. Am.
Chem. Soc. 1956, 78, 2514–2518.
27. Bo¨ker, M.; Jo¨rdening, H.-J.; Buchholz, K. Biotech. Bio-
eng. 1994, 43, 856–864.
47. Simiand, C.; Samain, E.; Martin, O. R.; Driguez, H.
Carbohydr. Res. 1995, 267, 1–15.
48. Reh, K.-D.; Jo¨rdening, H.-J.; Buchholz, K. Enzyme Eng.
1990, 10, 723–729.
28. Bourne, E. J.; Hartigan, J.; Weigel, H. J. Chem. Soc.
1961, 00, 1088–1092.
29. Iriki, Y.; Hehre, E. J. Arch. Biochem. Biophys. 1969, 134,
130–138.
30. Bailey, R. W.; Barker, S. A.; Bourne, E. J.; Grant, P. M.;
Stacey, M. J. Chem. Soc. 1958, 00, 1895–1902.
31. Bourne, E. J.; Hartigan, J.; Weigel, H. J. Chem. Soc.
1959, 00, 2332–2337.
32. Neely, W. B. Arch. Biochem. Biophys. 1959, 79, 154–161.
33. Suzuki, H.; Hehre, E. J. Arch. Biochem. Biophys. 1964,
105, 339–348.
49. Demuth, B.; Jo¨rdening, H.-J.; Buchholz, K. Biotech. Bio-
eng. 1999, 62, 583–592.
50. Heincke, K.; Demuth, B.; Jo¨rdening, H.-J.; Buchholz, K.
Enzym. Microbial Technol. 1999, 24, 523–534.
51. Reischwitz, A.; Reh, K.-D.; Buchholz, K. Enzym. Micro-
biol. Technol. 1995, 17, 457–461.
52. Stoppok, E.; Buchholz, K. The Production of 3-Keto
Derivatives of Disaccharides. In Methods in Biotechnol-
ogy—Carbohydrate Biotechnology Protocols; Bucke, C.,
Ed.; Humana Press: Totowa, NJ, 1999; pp 277–289.
53. Ferrier, R. J. J. Chem. Soc. 1964, 00, 5443.
54. Lichtenthaler, F. W.; Klotz, J.; Flath, F.-J. Liebigs Ann.
Chem. 1995, 00, 2069–2081.
34. Hehre, E. J.; Suzuki, H. Arch. Biochem. Biophys. 1966,
113, 675–683.
35. Bhattacharjee, M. K.; Mayer, R. M. Carbohydr. Res.
1993, 242, 191–201.
55. Bradbury, J. H.; Collins, J. G. Carbohydr. Res. 1979, 71,
15–24.
36. Figures, W. R.; Edwards, J. R. Carbohydr. Res. 1976, 48,
245–253.
37. Jung, S. M.; Mayer, R. M. Arch. Biochem. Biophys. 1981,
208, 288–295.
38. Grier, T. J.; Mayer, R. M. Arch. Biochem. Biophys. 1981,
212, 651–659.
39. Grier, T. J.; Mayer, R. M. Arch. Biochem. Biophys. 1981,
212, 651–659.
40. Binder, T. P.; Robyt, J. F. Carbohydr. Res. 1986, 154,
229–238.
41. Eklund, S. H.; Robyt, J. F. Carbohydr. Res. 1988, 177,
253–258.
42. Michiels, A. G.; Wang, A. Y.; Clark, D. S.; Blanch, H.
W. Appl. Biochem. Biotechnol. 1991, 31, 237–246.
43. Bhattacharjee, M. K.; Mayer, R. M. Carbohydr. Res.
1985, 142, 277–284.
56. Seymour, F. R.; Knapp, R. D.; Bishop, S. H. Carbohydr.
Res. 1976, 51, 179–194.
57. Seymour, F. R.; Knapp, R. D. Carbohydr. Res. 1980, 81,
67–103.
58. Remaud-Simeon, M.; Lopez-Munguia, A.; Pelenc, V. P.;
Paul, F. B.; Monsan, P. F. Appl. Biochem. Biotechnol.
1994, 44, 101–117.
59. Demuth, K. Ph.D. Thesis, Technical University of Braun-
schweig, Germany, 1999.
60. Demuth, K.; Jo¨rdening, H.-J.; Buchholz, K. In Food
Biotechnology; Bielecki, S.; Tramper, J.; Polak, J., Eds.;
Elsevier Science: Amsterdam, 2000; pp 123–135.
61. Ebert, K. H.; Patat, Z. F. Naturforschg. 1962, 17b, 738–
748.
62. Ebert, K. H.; Schenk, Z. G. Naturforschg. 1968, 23b,
788–798.