Characterization of Recombinant Sucrose Synthase 1 from Potato for the Synthesis of Sucrose Analogues

Article abstract of DOI:10.1002/1615-4169(200108)343:6/7<655::AID-ADSC655>3.0.CO;2-A

The characteristics and the application of recombinant sucrose synthase 1 (SuSy1) from potato for the synthesis of sucrose analogues are described. With UDP-Glc as donor substrate SuSy1 accepts a variety of ketoses, e.g., 1-deoxy-1-fluoro-D-fructose (6; 100%), L-sorbose (7, 55%), and D-xylulose (8; 42%), as well as aldoses, e.g., D-talose (15; 36%), D-idose (16; 24%), D-lyxose (12; 48%), L-arabinose (13; 36%), and D-ribose (14; 7%). Kinetic analyses revealed that the non-natural acceptors 6 (k_cat/ K_m = 3.5 s^-1 mM^-1), 7 (k_cat/K_m = 1.10^-2 s^-1 mM^-1), 8 (k_cat/K_m = 2.10^-2 s^-1 mM^-1), and 12 (k_cat/K_m = 2.10^-2 s^-1 mM^-1) were relatively poor substrates when compared to D-fructose (5; k_cat/K_m = 34.1 s^-1 mM^-1). It is concluded that the configuration and/or presence of the hydroxymethyl group at C5 determine the affinity of the ketoses for SuSy1. The acceptance of aldoses can be explained by their flexible chair conformations, which lead to isosteric hydroxy groups recognized by SuSy1. The preparative synthesis of sucrose analogues yielded 1′-deoxy-1′-fluoro-β-D-fructofuranosyl-α-D- glucopyranoside (1), [¹³C₁]-β-D-fructofuranosyl-α-D-glucopyranoside (2), α-D-glucopyranosyl-α-L-sorbofuranoside (3), and α-D-glucopyranosyl-α-D-lyxopyranoside (4), in a 0.1-1.0 g scale. The sucrose analogues 1, 3, and 4 were not hydrolyzed by invertase, which makes them valuable tools for studies on signal transduction pathways and sugar transport in plants.

Full text of DOI:10.1002/1615-4169(200108)343:6/7<655::AID-ADSC655>3.0.CO;2-A

FULL PAPER
Characterization of Recombinant Sucrose Synthase 1
from Potato for the Synthesis of Sucrose Analogues
Ulrike RoÈmer,a Nadja Nettelstroth,a Walter KoÈckenberger,b Lothar Ellinga,*
a Institute of Enzyme Technology, Heinrich-Heine-University DuÈ sseldorf, Research Center JuÈ lich, 52426 JuÈ lich, Germany
Fax: (+49) 2461
b Magnetic Resonance Centre, School of Physics and Astronomy, University of Nottingham, Nottingham NG7 2RD, UK
Fax: +(44) 115 846 6003, e-mail: Walter.Kockenberger@nottingham.ac.uk
-612-490, e-mail: L.Elling@fz-juelich.de
-
-
Received April 27, 2001; Accepted July 4, 2001
Abstract: The characteristics and the application of the affinity of the ketoses for SuSy1. The acceptance
recombinant sucrose synthase 1 (SuSy1) from potato of aldoses can be explained by their flexible chair
for the synthesis of sucrose analogues are described. conformations, which lead to isosteric hydroxy
With UDP-Glc as donor substrate SuSy1 accepts a groups recognized by SuSy1. The preparative syn-
variety of ketoses, e.g., 1-deoxy-1-fluoro-d-fructose thesis of sucrose analogues yielded 1'-deoxy-1'-
(6; 100%), l-sorbose (7, 55%), and d-xylulose (8; fluoro-b-d-fructofuranosyl-a-d-glucopyranoside
42%), as well as aldoses, e.g., d-talose (15; 36%), d- (1), [13C1]-b-d-fructofuranosyl-a-d-glucopyranoside
idose (16; 24%), d-lyxose (12; 48%), l-arabinose (13; (2), a-d-glucopyranosyl-a-l-sorbofuranoside (3),
36%), and d-ribose (14; 7%). Kinetic analyses re- and a-d-glucopyranosyl-a-d-lyxopyranoside (4), in
vealed that the non-natural acceptors 6 (kcat
/
a 0.1 ± 1.0 g scale. The sucrose analogues 1, 3, and 4
Km = 3.5 s±1 mM±1), 7 (kcat/Km = 1´10±2 s±1 mM±1), 8 were not hydrolyzed by invertase, which makes
(kcat/Km = 2´10±2 s±1 mM±1), and 12 (kcat/Km = 2´10±2 them valuable tools for studies on signal transduc-
s±1 mM±1) were relatively poor substrates when tion pathways and sugar transport in plants.
compared to d-fructose (5; kcat/Km = 34.1 s±1 mM±1).
It is concluded that the configuration and/or pre- Keywords: Enzyme catalysis; glycosides; glycocon-
sence of the hydroxymethyl group at C5 determine jugates; sucrose synthase; sucrose analogues
d-glucose,[6] a-d-glucopyranosyl-b-d-xylulo-furano-
Introduction
side, and a-d-glucopyranosyl-a-d-lyxo-pyranoside[7]
with SuSy purified from rice grains. Card et al.[8] used
The plant glycosyltransferase sucrose synthase (SuSy,
SuSy for the synthesis of fluoro and azido analogues of
EC 2.4.1.13) plays an important role in the metabo-
sucrose to study the binding specificity of a putative
lism of sucrose and its conversion to polysaccharides.
sucrose carrier protein.
SuSy catalyzes the cleavage of sucrose with nucleo-
The work presented in this paper comprises a de-
side diphosphates and provides the plant cell with ac-
tailed analysis of the substrate specificity spectrum
tivated sugar precursors for sucrose-starch transfor-
mation[1] and for cell wall synthesis.[2] SuSy
of recombinant sucrose synthase 1 from potato and
kinetic data of different acceptor substrates for the
represents a unique case among the Leloir glycosyl-
transferases by catalyzing in vitro both the cleavage
synthesis direction (Scheme 1). The sucrose anal-
o
gues, 1'-deoxy-1'-fluoro-b-d-fructofuranosyl-a-d-
of sucrose with nucleoside diphosphate yielding d-
fructose and nucleotide-activated d-glucose, and the
readily reversible reaction. It has been shown for de-
veloping potato tubers and other plant tissues that the
reaction is also reversible in vivo.[3] In contrast to
other enzymes of the sucrose metabolism, SuSy
shows a wide specificity for nucleoside diphosphates
in the sucrose cleavage direction.[4] In direction of su-
crose synthesis the enzyme shows also a wide specifi-
city.[5] This ability was utilized in the synthesis of the
sucrose analogues 2-O-6-deoxy-a-l-sorbofuranosyl-
glucopyranoside (1), [13C1]-b-d-fructofuranosyl-a-d-
glucopyranoside (2), a-d-glucopyranosyl-a-l-sorbo-
furanoside (3), and a-d-glucopyranosyl-a-d-lyxopyr-
anoside (4) were synthesized. All products were char-
acterized by H and 13C NMR. Work is in progress to
1
utilize the sucrose analogues for studies on sugar
sensing in plants[9] and on sucrose transport in plants
by in vivo NMR techniques (KoÈckenberger, in
preparation), respectively.
Adv. Synth. Catal. 2001, 343, No. 6±7
Ó WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2001
1615-4150/01/34306±7-655±661 $ 17.50-.50/0
655

asc.wiley-vch.de
Scheme 1. General scheme for the synthesis of sucrose analogues with recombinant sucrose synthase 1 from potato.
UDP-Glc (4-fold Km value) causes a lower Vmax value.
In comparison to enzyme preparations from potato,
Results and Discussion
recombinant SuSy1 shows a 3.5-fold to 13.5-fold high-
Ketoses as Acceptor Substrates
er affinity for UDP-Glc.[12] The Km value for 5 is com-
parable to those determined by Pressey[12b] and Mur-
Different ketoses were tested to explore the acceptor
ata[12c] for the potato enzyme; however, a substrate
substrate spectrum of SuSy1. Figure 1 illustrates that
inhibition for the donor and acceptor substrate was
not found. In contrast, a substrate inhibition above
20 mM of 5 was described for SuSy from maize.[13] Na-
kai et al.[14] found Km values of 7.72 mM and 0.4 mM
for 5 and UDP-Glc, respectively, for a recombinant su-
crose synthase from mung bean seedlings expressed
in E. coli. The kinetic analysis for 7 and 8 reveals,
however, that these ketoses are poor substrates with
an approx. 1000-fold decreased catalytic efficiency
(kcat/Km) for SuSy1 (Table 1). We conclude that the
configuration at C5 determines the affinity of the ke-
tose substrate.
the primary alcohol function at C1 of b-d-fructose (5)
can be replaced by a fluoro group in 6. Further varia-
tion of the ketoses reveals a crucial role for the con-
figuration at C3 and C4. The 3S,4R configuration in
l-sorbose (7) and d-xylulose (8) is important for the
ketose substrate to be recognized by SuSy1, whereas
d-tagatose (9), d-psicose (10), and d-sorbose (11) are
not accepted. However, the configuration and/or pre-
sence of the hydroxymethyl group at C5 appear not to
be crucial. The acceptance of 7 and 8 was also de-
scribed for sucrose synthase from green peas[10] and
sugar beet roots.[11] The kinetic constants for the ac-
cepted ketoses were determined to optimize the pre-
parative syntheses of sucrose analogues. They are
summarized in Table 1. Both natural substrates 5
Aldoses as Acceptor Substrates
and UDP-Glc show an inhibition at high substrate The idea to test also aldoses as acceptor substrates
concentrations. The relatively low KiS value of UDP- was derived from the fact that in aqueous solution
Glc implies that the donor substrate concentration the equilibrium form of b-d-fructose appears to be
limits the reaction rate of SuSy1. This is illustrated 70% pyranose and 23% furanose. The 1C4 chair con-
for 5 as a variable substrate, when already 2 mM formation is thermodynamically favored by the b-d-
Figure 1. Ketoses as acceptor substrates for recombinant SuSy1 from potato. The monosaccharides (10 mM) were tested
with the donor UDP-Glc (2 mM) in 50 mM HEPES buffer, pH 8.0, for 16 h at 30 °C. The formation of disaccharides and the
conversion of UDP-Glc was monitored by HPLC. The relative yields refer to the donor substrate UDP-Glc.
656
Adv. Synth. Catal. 2001, 343, 655±661

FULL PAPER
Table 1. Kinetic constants of recombinant SuSy1 from potato for the sucrose synthesis direction.
Substrate
Km [mM]
KiS [mM]
Vmax [U mg±1
]
kcat [s±1
]
kcat/Km [s±1 mM±1
]
UDP-Glc[a]
5[a]
0.46
2.05
6.28
226.5
244.3
125.7
2.30
35.88
21.76
10.63
3.29
0.35
0.61
0.35
142.8
69.8
21.7
2.3
4.0
2.3
310.4
34.1
3.5
0.01
0.02
0.02
6[a]
±
7[b]
521.2
8[b]
±
±
12[b]
[a] The activity assay was performed without addition of alkaline phosphatase.
[b] The activity assay was performed with addition of alkaline phosphatase.
strates. However, we found that only a few were read-
ily accepted. A common feature of the accepted al-
doses is that the defined anomers, as depicted in
Figure 2, have the flexible 4C1/1C4 chair conforma-
tions. Whether this feature represents an exclusive
criterion for SuSy1 cannot be established completely
by these investigations. The kinetic analysis (Table 1)
reveals 12 to be a poor substrate. However, in com-
parison to the ketoses 7 and 8 the lower Km value in-
dicates a better affinity of SuSy1. The comparison of
the accepted aldoses with the 1C4 chair conformation
of b-d-fructopyranose suggests that their conforma-
tional flexibility could contribute to isosteric config-
urations of their hydroxy groups, which are recog-
nized by SuSy1. Figure 3 illustrates our hypothesis
for 12 and 13, which predicts also the stereoselective
product formation. First evidence for our hypothesis
is reported below for the preparative synthesis of a
sucrose analogue, where 12 is converted by SuSy1 to
yield the non-reducing disaccharide 4 (Figure 4).
Figure 2. Aldoses as acceptor substrates for recombinant
SuSy1 from potato. The monosaccharides (10 mM) were
tested with the donor UDP-Glc (2 mM) in 50 mM HEPES buf-
fer, pH 8.0, for 16 h at 30 °C. The formation of disaccharides
and the conversion of UDP-Glc was monitored by HPLC. The
relative yields refer to the donor substrate UDP-Glc.
fructopyranose. We hypothesized that SuSy1 may re-
cognize the b-d-fructopyranose conformation rather
than removing the b-d-fructofuranose from the equi-
librium during the synthesis reaction. These led us to
test more than 20 different aldoses as acceptor sub-
Enzymatic Synthesis of Sucrose Analogues
Figure 4 and Table 2 summarize the synthesis of the
sucrose analogues 1'-deoxy-1'-fluoro-b-d-fructofura-
nosyl-a-d-glucopyranoside (1), [13C1]-b-d-fructofura-
Figure 3. Hypothetical model to demonstrate the isosteric configurations of the hydroxy groups of a-d-lyxose and b-l-arabi-
nose in comparison to the 1C4 chair conformation of b-d-fructose. Isosteric hydroxy groups are highlighted. Putative sites for
the transfer of the donor substrate are indicated by arrows. For a-d-lyxose the model product was confirmed by preparative
synthesis and characterization by NMR analysis.
Adv. Synth. Catal. 2001, 343, 655±661
657

asc.wiley-vch.de
Table 3. Assay of sucrose analogues as substrates of inver-
tase.
Product
Relative activity [%]
1[a]
2
3
100
0.04
0.06
0.12
4
[a] Unlabeled sucrose.
3 in an analytical scale (16.5 mg) using 5 U enzyme
with an overall yield of only 7.5%. This synthesis was
improved by taking into account the kinetic data and
by using both a fed-batch-technique and an addition
of alkaline phosphatase. Peters et al.[6] synthesized 2-
O-6-deoxy-a-l-sorbofuranosyl-a-d-glucopyranoside
with an overall yield of 17% (50 mg) using 8 U SuSy
from rice grains.
In studies on sugar signaling[9,16] and sugar trans-
port[17] in plants it is sometimes advantageous if su-
crose analogues are not hydrolyzed by invertases.
Such analogues can be used to determine unambigu-
ously the response between the signaling compound
and the receptor. Table 3 demonstrates that the prod-
ucts 2, 3, and 4 are not hydrolyzed by invertase.
Therefore, the very low residual activity of invertase
for these sucrose analogues makes them valuable
tools for further biochemical studies.
Figure 4. Products of the enzymatic synthesis with recombi-
nant SuSy1: 1-deoxy-1-fluorofructofuranosyl-a-d-glucopyr-
anoside (1),
[
13C1]-b-d-fructofuranosyl-a-d-glucopyrano-
side (2), a-d-glucopyranosyl-a-l-sorbofuranoside (3), and
a-d-glucopyranosyl-a-d-lyxopyranoside (4).
nosyl-a-d-glucopyranoside (2), a-d-glucopyranosyl-
a-l-sorbofuranoside (3), and a-d-glucopyranosyl-a-
d-lyxopyranoside (4). The repetitive use of all en-
zymes during the syntheses (repetitive-batch mode)
resulted in efficient enzyme productivities and
space-time yields (Table 2). Moreover, these data also
reflect the catalytic efficiency of SuSy1 with reference
to the acceptor substrates. In comparison with the ke-
tose 7 the affinity of SuSy1 for the aldose 12 is higher
and therefore results in higher enzyme productivity.
This is also evident in the case of the [13C1]-labeled
natural substrate 5 and the derivative 6. In compari-
son to previous published synthesis we could improve
the overall yield for the products 1 and 2. Card and
Hitz[8a] synthesized 1 with SuSy from barley seeds
with an overall yield of 59% (507 mg). Duker and Ser-
ianni[15] synthesized 2 with an overall yield of 25%
(85.6 mg) using 50 U SuSy. With sucrose synthase
purified from rice grains Grothus et al.[7] synthesized
Conclusion
The presented results provide for the first time details
of the structural requirements for acceptor substrates
to be recognized by recombinant SuSy1 from potato.
Although some compounds acted as poor substrates
the synthesis yielded unique sucrose analogues for
biochemical studies of signal transduction and sugar
transport in plants. Further information is still re-
quired to investigate the donor substrate spectrum of
this enzyme. However, already it can be concluded
that SuSy1 represents a unique biocatalyst for use in
carbohydrate engineering.
Table 2. Sucrose analogues synthesized with recombinant SuSy1 from potato.
Product
Synthesis yield [%]
Enzyme productivity[a] [mg U±1
]
Space-time yield [g L±1 day±1
]
Overall yield [%] Amount [g]
1
2
3
4
100[b]
94[b]
81[c]
81[c]
25
61
3.7
7.1
1.68
3.03
0.57
1.60
85[b]
55[b]
35[c]
69[c]
0.86
0.71
0.079
0.160
[a] Milligrams of formed product per unit of recombinant SuSy1.
[b] With reference to the acceptor substrate.
[c] With reference to the donor substrate.
658
Adv. Synth. Catal. 2001, 343, 655±661

FULL PAPER
Glc (1 mM) as donor substrate. The assay mixtures contain-
ing 200 mU mL±1 SuSy1 and 1 U mL±1 alkaline phosphatase
were incubated in 200 mM HEPES buffer, pH 8.0, for 16 h at
30 °C. The reaction was stopped by heating for 5 min at
95 °C. The formation of disaccharides was monitored by
HPLC analysis using an Aminex HPX-87C column (300 ´
7.8 mm, BioRad, Munich) and elution with distilled water at
85° C. The formation of UDP, UMP, and uridine was also ob-
served by HPLC.[19]
Experimental Section
Recombinant Sucrose Synthase 1 from Potato
(Solanum tuberosum)
The gene sus1 from Solanum tuberosum L., which encodes
SuSy1, was cloned in the expression plasmid pDR195 under
the control of the proton ATPase (PMA1) promoter. The plas-
mid was transformed in the Saccharomyces cerevisiae strain
22574d. The recombinant yeast cells expressing SuSy1 con-
stitutively during growth were cultivated on a 30-L scale
yielding 796 g wet cells. From 200 g homogenized cells
874 U SuSy1 with a yield of 81% were obtained by a partial
purification step with anion exchange chromatography.
The enzyme fraction was free of contaminating invertases
and phosphatases. With reference to the SuSy1 activity
0.05% of hydrolases cleaving nucleotide sugars, and 13%
hexokinase as well as 0.05% phosphoglucose isomerase
were present. The partially purified SuSy1 fraction was used
for the synthesis of nucleotide-activated sugars and sucrose
analogues. Kinetic studies required further purification by
immobilized metal ion chromatography (IMAC) giving an
86% yield. The activities of contaminating enzymes could
be minimized to 10% of the activities reported above.
Synthesis of 1-Deoxy-1-fluorofructofuranosyl-a-d-
glucopyranoside (1)
1-Deoxy-1-fluoro-d-fructose (6) was obtained by the reaction
described by Card and Hitz.[8a] In brief, the readily available
2,3:4,5-di-O-isopropylidene-d-fructopyranose was converted
into the triflate using the procedure described by Binkley et
al.[20] The triflate was fluorinated by TASF [tris(dimethylami-
no)sulfur(trimethylsilyl) difluoride, Aldrich Chemicals] in re-
fluxing tetrahydrofuran. After removal of the isopropylidene
protection groups, 1-deoxy-1-fluoro-d-fructose was obtained
as a syrup in 75% yield. The synthesis of 1'-deoxy-1'-fluorosu-
crose was carried out by the repetitive-batch technique.[21]
The reaction mixture (100 mL) containing 0.96 mmol 1-
deoxy-1-fluorofructose (176 mg) and 1 mmol UDP-a-d-glu-
cose in 200 mM HEPES buffer, pH 8.0, was gently stirred at
30 °C after adding of 40 U recombinant SuSy1 and 200 U alka-
line phosphatase (Roche Diagnostics, Mannheim). The
course of the reaction was monitored by HPLC analysis of
the product with an Aminex HPX-87C column as described
above. After 48 h the enzymes were recovered by ultrafiltra-
tion and used in a second and third batch, after the addition
of new substrates. The total yield of the combined product so-
lutions was 2.9 mmol (100%) for 1'-deoxy-1'-fluorosucrose.
In a first purification step the product solution was adjusted
to pH 8.6 and loaded onto an anion exchanger column
(HCOO± form) filled with AG 1-X8 resin (100 ± 200 mesh,
122 mL bed volume, BioRad, Munich), which was equili-
brated with distilled water. Elution with distilled water (linear
velocity: 56.5 cm h±1) resulted in a product pool, which was
concentrated by vacuum evaporation to a final volume of
5 mL. The disaccharide was further purified by chromatogra-
phy on an AG 50 W-X8 resin column (200 ± 400 mesh, Ca2+
form, 1532 mL bed volume, BioRad, Munich). Elution with
distilled water (linear velocity: 3 cm h±1) gave the fractions
containing the disaccharide, which were pooled and lyophi-
lized. The dry product was dissolved in 10 mL absolute
methanol and crystallized at 25 °C. The product 1 was ob-
tained as a white solid in an overall yield of 85% correspond-
ing to 2.5 mmol (860.5 mg) with an HPLC purity of 89%. NMR
spectroscopy (Bruker, DAX500) (11.7 T) of 1 revealed the
typical couplings between 19F and 1H or 13C. 1H NMR
Enzyme Assay
Enzyme activity of SuSy1 was measured in the direction of
sucrose cleavage. In a total volume of 1 mL of 200 mM
HEPES buffer, pH 7.6, recombinant SuSy1 from potato was
incubated with 2 mM UDP (Sigma, Deisenhofen) and
500 mM sucrose for 10 min at 30 °C.[18] The reaction was
stopped by heating at 95 °C for 5 min. The formation of
UDP-a-d-glucose (UDP-Glc) was monitored by HPLC.[19]
One unit is defined as the amount of enzyme for the forma-
tion of 1 lmol UDP-Glc per min under standard conditions.
Kinetic Data
All kinetic measurements in the synthesis direction were
performed in 200 mM HEPES buffer, pH 8.0 at 30 °C. The ki-
netic constants for UDP-Glc (Sigma, Deisenhofen) and the
acceptors 5 and 6 were obtained at a constant concentrations
of 10 mM for the natural acceptor 5 and a constant concen-
tration of 2 mM for the donor substrate UDP-Glc. The kinetic
data 7, 8, and 12 were determined at a constant concentra-
tion of 1 mM UDP-glucose and in the presence of 1 U mL±1
alkaline phosphatase (Roche Diagnostics, Mannheim). Initi-
al rate measurements were obtained by allowing a maxi-
mum conversion of 10% of the variable substrate followed
by HPLC detection[19] of formed UDP or UMP and uridine
when using alkaline phosphatase. The kinetic constants
were obtained by a non-linear regression analysis of the data
using the Michaelis-Menten equation: V = (Vmax ´ S)/(S +
Km) or the kinetic equation for substrate inhibition: V =
(Vmax ´ S)/(S + Km + S2/KiS).
(500 MHz, D2O): d = 4.39 (1'-H, m, JH,F = 46.6 Hz, JH,H
=
10.4 Hz); 19F NMR (470 MHz, D2O, CFCl3): d = ±229.4 (m,
JF,H = 46.6 Hz); 13C NMR (125 MHz, D2O): d = 80.7 (1'-C, d,
JC,F = 174.2 Hz), 101.8 (2'-C, d, JC,F = 19.6 Hz).
Synthesis of [13C1]-b-d-Fructofuranosyl-a-d-
glucopyranoside (2)
Variation of Acceptor Substrates
Different ketoses and aldoses (10 mM) were tested as accep-
tor substrates in the synthesis reaction of SuSy1 with UDP-
The synthesis was also carried out with the repetitive-batch
technique. The reaction mixture (100 mL) containing
Adv. Synth. Catal. 2001, 343, 655±661
659

asc.wiley-vch.de
1 mmol [13C1]-b-d-fructofuranoside (Euriso-top, Gif-Sur-Yv-
ette, France) and 1 mmol UDP-Glc in 200 mM HEPES buffer,
pH 8.0, was gently stirred at 30 °C after the addition of 20 U
recombinant SuSy1 from potato and 200 U alkaline phospha-
tase. The course of reaction was controlled by HPLC. After
24 h the enzymes were recovered by ultrafiltration and used
in the three successive batches, together with new sub-
strates. The yield of the combined product solutions was
3.78 mmol (94%) for 2. Prior to isolation the product solution
was adjusted to pH 8.6 and then loaded onto an anion ex-
changer column (Cl± form) filled with Dowex 1x2 resin (100
± 200 mesh, 400 mL bed volume, Serva, Heidelberg), which
was equilibrated with distilled water. Elution with distilled
water (linear velocity: 22 cm h±1) resulted in a product pool,
which was concentrated by vacuum evaporation to a final
volume of 16 mL. The disaccharide (aliquots of 2 mL prod-
uct solution) was further purified by chromatography on a
Bio-Gel P2 resin column (extra fine, 500 mL bed volume,
BioRad, Munich). Elution with distilled water (linear velo-
city: 5 cm h±1) gave the disaccharide-containing fractions,
which were pooled, combined, and lyophilized. The product
2 was obtained in an overall yield of 55% corresponding to
2.07 mmol (712 mg) with an HPLC purity of 89%. NMR spec-
troscopy (Bruker, DPX 400 advance series) revealed the typi-
cal signals and couplings. 13C NMR (100 MHz, D2O): d = 62.2
ing 9 mmol of 12 and 0.045 mmol UDP-Glc in 200 mM
HEPES buffer, pH 8.0, was gently stirred at 30 °C after addi-
tion of 26.5 U recombinant SuSy1 and 115 U alkaline phos-
phatase. The course of reaction was monitored by HPLC as
described before. After incubation for 1.5 h 0.684 mmol
UDP-Glc was added to the reaction mixture over 38 h. After
the batch had been stirred for another 9 h, the reaction was
terminated by separation of the enzymes with ultrafiltration.
The synthesis yield was 0.59 mmol (81%) for 4. Product iso-
lation was carried out as described for 3. For the sucrose
analogue 4 an overall yield of 69% (0.50 mmol, 159.9 mg)
was obtained. 1H NMR (500 MHz, D2O): d = 509 (1-H, d);
4.99 (1'-H, d), 3.89 (2'-H, t), 3.87 (3'-H, t), 3.79-3.85 (5'-Ha,b,
3-H, 5-H, 6-Ha,b, m), 3.68 (4'-H, ddd), 3.55 (2-H, t), 3.38 (4-H,
3
3
3
3
3
t); J1,2 = 3.8 , J1',2' = 2.8, J2',3' = 3.0, J3',4' = 2.2, J3,4 = 9.4,
4J1',3' = 12.3 Hz; 13C NMR (125 MHz, D2O): d = 95.72 (C-1'),
94.18 (C-1), 73.00 (C-3), 72.77 (C-5), 71.25 (C-2), 70.77 (C-
3'), 69.95 (C-2'), 69.93 (C-4), 67.03 (C-4'), 63.23 (C-5'), 60.87
(C-6). ESI-MS (Finnigan MAT, LCQ): found: 311.0 m/z for [M
± H]±, calculated: 311.11 m/z for [M ± H]±.
Invertase Assay
The sucrose analogues 1, 3, and 4 were tested as substrates
of yeast invertase (Sigma, Deisenhofen) and compared to
the natural substrate sucrose. Invertase catalyzes the hydro-
lysis of sucrose, which results in the formation of d-glucose
and d-fructose. The photometric invertase assay described
by Bergmeyer and Bernt[22] was used. In summary, d-glu-
cose was measured through a conversion with glucose ox-
idase. The concomitant formation of hydrogen peroxide
was coupled to the peroxidase-catalyzed oxidation of the
dye ABTS [2,2'-azinobis(3-ethylbenzthiazoline)-6-sulfo-
nate]. In detail, the assay was carried out by incubation of
10 U ml±1 invertase (Sigma, Deisenhofen), 9 U ml±1 glucose
oxidase (Sigma, Deisenhofen), and 1.5 U ml±1 peroxidase
(Merck, Darmstadt) in 0.2 M phosphate buffer (200 ll),
pH 6.0 at 25 °C, containing 0.0125 mM sucrose or sucrose
analogues, and 0.5 mg ml±1 ABTS. The change of absorption
was observed at 405 nm giving initial rate measurements as
DE min±1. The initial rate of invertase for sucrose was set to
100%.
(s, C1'); 1H NMR (400 MHz, D2O): d = 3.56 (1'-H, dd, JHCË
144 Hz), 5.33 (1-H, dd, JHCË = 169 Hz).
=
Synthesis of a-d-Glucopyranosyl-a-l-
sorbofuranoside (3)
The synthesis was carried out in a final volume of 75 mL
using the fed-batch technique. The reaction mixture con-
taining 22.5 mmol 7 and 0.125 mmol UDP-Glc in 200 mM
HEPES buffer, pH 8.0, was gently stirred at 30 °C after the
addition of 50 U recombinant SuSy1 from potato and 225 U
alkaline phosphatase. The course of reaction was controlled
by HPLC as described above. After an incubation time of 19 h
0.4 mmol UDP-Glc was fed to the reaction mixture over the
next 48 h. After the batch has been stirred for another 25 h
the enzymes were separated by ultrafiltration. The synthesis
yield was 0.51 mmol (81%) of 3. The first purification step
was carried out as described above for 1. The disaccharide
was further purified by chromatography on a preparative
HPLC column [carbohydrate-Ca column (300 ´ 20 mm, CS,
Langerwehe) with elution by distilled water (flow rate:
1.2 mL min±1). The pooled and lyophilized fractions gave 3
in an overall yield of 35% (0.220 mmol, 79 mg). 1H NMR
(500 MHz, D2O): d = 5.31 (1-H, d), 4.40 (4'-H, t), 4.31 (5'-H,
Acknowledgements
L.E. acknowledges financial support by the Deutsche For-
schungsgemeinschaft, DFG, (SFB380, Teilprojekt B26) and
the Fonds der Chemischen Industrie. W.K. is grateful for the
support and advice of Drs C. Roby, J. Defaye, and A. Gadelle
(CEA Grenoble) during the synthesis of fluorofructose.
dq), 4.14 (3'-H, d), 3.61
-3.82 (1'-Ha,b, 3-H, 5-H, 6-Ha,b, m),
3
3
3
3.50 (2-H, dd), 3.34 (4-H, t); J1,2 = 4.1, J3',4' = 7.3, J4',5'
=
4
7.6, J2,4 = 10.0 Hz; 13C NMR (125 MHz, D2O): d = 104.03 (C-
2'), 92.66 (C-1), 78.59 (C-5'), 77.74 (C-3'), 75.13 (C-4'), 73.11
(C-3), 73.05 (C-5), 71.48 (C-2), 69.84 (C-4), 61.06 (C-6'), 60.97
(C-1'), 60.81 (C-6). ESI-MS (Finnigan MAT, LCQ): found:
341.1 m/z for [M ± H]±, calculated: 341.12 m/z for [M ± H]±.
References
[1] R. Zrenner, M. Salanoubat, L. Willmitzer, U. Sonne-
wald, Plant J. 1995, 7, 97±107.
[2] Y. Amor, C. H. Haigler, S. Johnson, M. Wainscott, D. P.
Delmer, Proc. Natl. Acad. Sci. USA 1995, 92, 9353±
9357.
Synthesis of a-d-Glucopyranosyl-a-d-
lyxopyranoside (4)
The synthesis of 4 was performed in a final volume of 58 mL
with the fed-batch technique. The reaction mixture contain-
[3] P. Geigenberger, M. Stitt, Planta 1993, 189, 329±339.
660
Adv. Synth. Catal. 2001, 343, 655±661

FULL PAPER
[4] H. G. Pontis, in International Review of Biochemistry,
Plant Biochemistry II, Vol. 13, (Ed.: D. H. Northcote),
University Park Press, Baltimore, 1977, pp. 79±117.
[5] (a) L. Elling, M. Grothus, M.-R. Kula, Glycobiology
1993, 3, 349±355; (b) R. C. Bean, W. Z. Hassid, J. Am.
Chem. Soc. 1955, 77, 5737±5738.
[13] D. C. Doehlert, Plant Sci. 1987, 52, 153±157.
[14] T. Nakai, N. Tonouchi, T. Tsuchida, H. Mori, F. Sakai,
T. Hayashi, Biosci. Biotech. Biochem. 1997, 61, 1500±
1503.
[15] J. M. Duker, A. S. Serianni, Carbohydr. Res. 1993, 249,
281±303.
[6] J. Peters, H.-P. Brockamp, T. Minuth, M. Grothus, A.
Steigel, M.-R. Kula, L. Elling, Tetrahedron: Asymmetry
1993, 4, 1173±1182.
[16] T. Roitsch, Curr. Opin. Plant Biol. 1999, 2, 198±206.
[17] J. M. Ward, C. Kuhn, M. Tegeder, W. B. Frommer, Int.
Rev. Cytol. 1998, 178, 41±71.
[7] M. Grothus, A. Steigel, M.-R. Kula, L. Elling, Carbo-
hydr. Lett. 1994, 1, 83±88.
[8] (a) P. J. Card, W. D. Hitz, J. Am. Chem. Soc. 1984, 106,
5348±5350; (b) P. J. Card, W. D. Hitz, K. G. Ripp, J. Am.
Chem. Soc. 1986, 108, 158±161.
[18] A. Zervosen, U. RoÈmer, L. Elling, J. Mol. Catal. B: En-
zymatic 1998, 5, 25±28; A. Zervosen, L. Elling, in Meth-
ods in Biotechnology: Carbohydrate Biotechnology Pro-
tocols, Vol. 10, (Ed.: C. Bucke), Humana Press,
Totowa, NJ, 1999, pp. 235±254.
[9] A. K. Sinha, U. RoÈmer, W. KoÈckenberger, M. Hofmann,
L. Elling, T. Roitsch, Plant J., submitted.
[10] R. C. Bean, W. Z. Hassid, J. Am. Chem. Soc. 1955, 77,
5737±5738.
[11] G. Avigad, Y. Milner, in Meth. Enzymol., Vol. VIII,
(Eds.: E. F. Neufeld, V. Ginsburg), Academic Press,
New York, 1966, pp. 341±345.
[12] (a) E. Slabnik, R. B. Frydman, C. E. Cardini, Plant Phy-
siol. 1968, 43, 1063±1068; (b) R. Pressey, Plant Physiol.
1969, 44, 759±764; (c) T. Murata, Agr. Biol. Chem.
1972, 36, 1815±1818.
[19] T. Ryll, R. Wagner, J. Chromatogr. 1991, 570, 77±88.
[20] R. W. Binkley, M. B. Ambrose, D. G. Hehemann, J.
Org. Chem. 1980, 45, 4387±4391.
[21] U. Kragl, A. GoÈdde, C. Wandrey, W. Kinzy, J. J. Cap-
pon, J. Lugtenburg, Tetrahedron: Asymmetry 1993, 4,
1193±1202.
[22] H. U. Bergmeyer, E. Bernt, in Methoden der enzyma-
tischen Analyse, Vol. 2, (Ed.: H. U. Bergmeyer), Verlag
Chemie, Weinheim, 1974, pp. 1250±1259.
Adv. Synth. Catal. 2001, 343, 655±661
661