Fructansucrase enzymes and sucrose analogues: A new approach for the synthesis of unique fructo-oligosaccharides

Abstract

Fructansucrase enzymes of lactic acid bacteria use the cheap compound sucrose (Glc-Fru) to synthesize a variety of poly- and oligosaccharide products. Recently, it has been shown that a variety of sucrose analogues (e.g. Gal-Fru, Man-Fru) can be efficiently synthesized. This has exciting potential for the development of novel (fructo) oligosaccharide derivatives.

Full text

ORIGINAL ARTICLE
Fructansucrase enzymes and sucrose analogues: A new approach for
the synthesis of unique fructo-oligosaccharides
S. KRALJ
1
, K. BUCHHOLZ
2
, L. DIJKHUIZEN
1
, & J. SEIBEL
2
1
Centre for Carbohydrate Bioprocessing, TNO-University of Groningen and Department of Microbiology, Groningen
Biomolecular Sciences and Biotechnology Institute (GBB), University of Groningen, Haren, The Netherlands and
2
Technical
Chemistry, Department for Carbohydrate Technology, Technical University Braunschweig, Braunschweig, Germany
Abstract
Fructansucrase enzymes of lactic acid bacteria use the cheap compound sucrose (GlcFru) to synthesize a variety of poly-
and oligosaccharide products. Recently, it has been shown that a variety of sucrose analogues (e.g. GalFru, ManFru) can
be efficiently synthesized. This has exciting potential for the development of novel (fructo) oligosaccharide derivatives.
Introduction
Oligosaccharides (OSs) and polysaccharides (PSs)
are ubiquitous in nature as components of a broad
range of molecular structures. They function as
structural scaffolds for energy storage, and as key
components in many biological processes. OSs and
PSs are associated with cell-cell communication,
differentiation, development, bacterial and viral
infections, and they are diagnostic tools in certain
cancers (Kaur et al. 2002; Nakakuki 2002; Seibel
et al. 2006b; Wong 2005). Due to the difficulty in
producing sufficient quantities at reasonable prices,
OSs and PSs are not commonly used in the
pharmaceutical industry (Rastall et al. 2003). Su-
crose chemistry has elicited much interest since
sucrose (a-
D
-glucopyranosyl b-
D
-fructofuranoside)
is available on a large scale with high purity at low
cost. A large variety of OSs and PSs can be
synthesized enzymatically from sucrose using glu-
cansucrases (GSs; Monchois et al. 1999) and
fructansucrases (FSs; Cerning 1990). These include
glucans containing various glucosidic linkages, e.g.
dextran with an a-1,6-bound glucose (Glc) back-
bone and a-1,2 and a-1,3 side chains (Binder et al.
1983; Buchholz et al. 2003; Coˆte´ & Robyt 1982;
Hehre 1941; Monchois et al. 1997; Remaud-Simeon
et al. 1994), alternan (a-1,6- and a-1,3-bound Glc
units ; Coˆ te´ & Robyt 1982), reuteran (a-1,4 and
a-1,6-bound Glc units; Kralj et al. 2002, 2004), and
the fructans (Yun 1996) levan with b-2,6-bound
fructose (Fru) units (Chambert et al. 1974; van
Hijum et al. 2001) and inulin (b-2,1-bound Fru
units; van Hijum et al. 2002, 2006).
The challenge of synthesis
Carbohydrate biosynthesis and glycosylation are
topics of current interest in academia and industry,
e.g. for OS and PS use in food ingredients, neu-
traceuticals and pharmaceuticals, such as vaccines,
glycoprotein therapeutics and glycosylated drugs in
general (Wong 2005). Glycosylation is important
with respect to pharmacokinetics and pharmacody-
namics, immunogenicity and efficiency of drugs
(Werner et al. 2007). Furthermore, glycosylation
may improve such properties as specificity, solubi-
lity, toxic effects and bitterness of neutraceuticals,
isoflavones and aroma peptides.
The potential of carbohydrates as functional foods
and new therapeutics has highlighted the require-
ment for a general availability of larger amounts of
various carbohydrate structures (Nakakuki 2002;
Rastall & Hotchkiss 2003). Thus, the synthesis of
OSs, PSs and glycosylated substances is a major
challenge. In this sense, activated sugars are key
substrates for synthesis and glycosylation (Seibel
et al. 2006b).
Correspondence: J. Seibel, Technical Chemistry, Department for Carbohydrate Technology, Technical University Braunschweig, Hans-
Sommer Strasse 10, 38106 Braunschweig, Germany. E-mail: j.seibel@tu-bs.de
Biocatalysis and Biotransformation, JanuaryFebruary 2008; 26(12): 3241
ISSN 1024-2422 print/ISSN 1029-2446 online #2008 Informa UK Ltd
DOI: 10.1080/10242420701789478

Fructansucrase enzymes
Most bacterial FSs known are levansucrases (LSs;
EC 2.4.1.10), synthesizing fructan PSs composed of
b-2,6-linked fructose units (levan; Gross et al. 1992;
Steinmetz et al. 1985; van Hijum et al. 2004); only
limited information is available about bacterial
inulosucrases (ISs; EC 2.4.1.9) producing b-2,
1-linked fructan PSs (inulin; Baird et al. 1973;
Olivares-Illana et al. 2002; Rosell & Birkhed 1974;
van Hijum et al. 2002).
LSs are present in Gram-positive, as well as
Gram-negative bacteria. However, ISs have been
identified exclusively in lactic acid bacteria (LAB;
van Hijum et al. 2006). Recently, more FS genes
have become available due to elucidation of several
LAB genome sequences. Interestingly, unique chi-
meric LS and IS enzymes have also been character-
ized recently, carrying additional (glucan-binding)
domains also found in GSs (Morales-Arrieta et al.
2006; Olivares-Illana et al. 2002).
Bacterial FSs cleave the glycosidic bond of their
natural substrate sucrose and use the energy released
to synthesize fructan PSs. Besides this polymeriza-
tion reaction, the enzymes also perform a hydrolysis
reaction, and in the presence of suitable acceptors
OSs are synthesized (Chambert et al. 1974; Seibel
et al. 2006c; van Hijum et al. 2006).
Mechanistic insights into LSs and ISs
LSs and ISs belong to glycoside hydrolase family 68
(GH68) and together with GH32 (e.g. invertase,
inulinase) they comprise clan GH-J, according to
CAZy (CAZy 2006, www.cazy.org). Two different
LS X-ray structures have been elucidated, the first
from the gram-positive bacterium Bacillus subtilis
(SacB, also with bound sucrose; Meng & Fu
¨tterer
2003) and one from the gram-negative bacterium
Gluconobacter diazotrophicus (Martinez-Fleites et al.
2005). At present, no 3D structural information is
available for an IS protein. In addition, three 3D
structures of GH32 representatives are now avail-
able: invertase from Thermotoga maritima (also with
the trisaccharide raffinose in the active site), fructan
1-exohydrolase from Cichorum intybus and exo-
inulinase from Aspergillus niger (Alberto et al. 2004,
2006; Nagem et al. 2004; Verhaest et al. 2005). All
five proteins possess a five-bladed b-propeller fold
with a deep, negatively-charged, central cavity. Their
active sites are positioned at the end of this cavity
with a funnel-like opening towards the molecular
surface.
The GH68 B. subtilis SacB X-ray structure with
sucrose trapped in the active site clearly shows
binding of sucrose in subsites 1 (fructosyl residue)
and 1 (glucosyl residue; Meng et al. 2003; subsite
nomenclature according to Davies et al. 1997). The
geometry of its active site shows a salt bridge
between E342 and R246, further disrupting subsites
2 and 3 (Meng & Fu
¨tterer 2003; Ozimek et al.
2004). A similar situation exists in Neisseria poly-
saccharea amylosucrase (sucrose active enzyme, but
belonging to family GH13; Skov et al. 2002).
Experimental data, indicating that FS enzymes per-
form a disproportionation type of reaction, corrobo-
rated the presence of only one sugar binding donor
subsite 1 in FS enzymes (Ozimek et al. 2006b).
Amino acid sequence alignments of various LSs
and ISs, in combination with site-directed mutagen-
esis experiments and structural data of family
members, pointed towards various relevant func-
tional amino acid residues, and provided the first
insights into structure-function relationships (Meng
& Fu
¨tterer 2003; Martinez-Fleites et al. 2005;
Ozimek et al. 2006a; Homann et al. 2007). In LSs
and ISs from various bacteria Asp
95
, Glu
352
and
Asp
257
(Bacillus megaterium numbering) constitute
the catalytic triad (Ozimek et al. 2004; Homann
et al. 2007). Most likely, the acid/base catalyst
Glu
352
protonates the glycosidic bond of sucrose,
while the Asp
95
residue acts as a nucleophile
(Chambert & Gonzy-Treboul 1976), attacking the
glucopyranosyl residue of the sucrose, presumably
forming an enzyme-fructosyl intermediate with in-
version of the glycosidic bond in the intermediate
state. The residue Asp
257
may assist in coordinating
the hydrogen bonds of the fructofuranoside at
positions 3-OH and 4-OH.
Poly- versus oligosaccharide synthesis
Although LSs and ISs use a similar substrate, these
enzymes differ strongly in their reaction and product
specificities, i.e. in b-2,6 vs. b-2,1 glycosidic bond
specificity (resulting in levan and inulin synthesis,
respectively), and in the ratio of hydrolysis vs.
transglycosylation activities (FOSs; Ozimek et al.
2006b).
The ratio between PS and OS synthesis activities
differs significantly depending on the enzyme
source. The LS of B. subtilis catalyses the formation
of a high-molecular-mass levan without transient
accumulation of oligofructan molecules (Chambert
et al. 1974; Hernandez et al. 1995). A similar
reaction was observed for the LS of Lactobacillus
reuteri 121 (Ozimek et al. 2006b) and the LS of B.
megaterium (Homann et al. 2007). This suggested
that the growing polymer chain remained bound to
the enzyme and that fructan chain elongation
proceeded via a processive type of reaction. In
contrast, the LS of G. diazotrophicus,Zymomonas
Fructansucrase enzymes and sucrose analogues 33

mobilis, and Lactobacillus sanfranciscensis, and the IS
of L. reuteri 121, mainly synthesized short FOSs;
kestose and nystose) from sucrose (Doelle et al.
1993; Hernandez et al. 1995; Korakli et al. 2001,
2003; Ozimek et al. 2006b, 2006a, b). These
enzymes thus employ a non-processive type of
reaction, involving release of the fructan chain after
(virtually) each fructosyl transfer.
Active site architecture (subsites

1, 1
and 2)
Amino acid residues located at subsites 1 and 1 in
the FS active site directly interacting with sucrose
have been identified based on the 3D structure of B.
subtilis SacB levansucrase with bound sucrose (Meng
& Fu
¨tterer 2003), and by several FSs mutagenesis
studies (Chambert & Petit-Glatron 1991; Homann
et al. 2007; Ozimek et al. 2004; Yanase et al. 2002).
A total of eight residues involved in catalysis and/or
interacting with the bound fructose and glucose at
subsites 1 (W85, D86, R246, D247 and W163) and
1 (R360, E340, and R246; B. subtilis SacB
numbering) are conserved in FSs. The FS subsite
1 is highly specific for binding fructose units
(Chambert et al. 1974; Hernandez et al. 1995;
Song et al. 1999), whereas subsite 1 is more
flexible, exhibiting affinity for glucose in binding
sucrose and raffinose, affinity for fructose in binding
sucrose, as an acceptor substrate during transglyco-
sylation, and affinity for different glycopyranosides
(mannose, galactose, fucose and xylose) and dis-
accharides (maltose, lactose and melibiose; Seibel
et al. 2005, 2006c). Some first insights regarding
products formed have been obtained. Key amino
acid residues in LSs important for OS formation
have been identified. For instance, changes in Arg
360
from B. subtilis SacB (Chambert & Petit-Glatron
1991) at subsite 1, and the similar amino acid
residue Arg
370
of B. megaterium SacB, resulted
in time-dependent accumulation of different OSs
during their catalysis, and accumulation of neokes-
tose (b-Fruf-2,6-a-Glcp-1,2-b-Fruf) and blastose
(b-Fruf-2,6-a-Glcp; Homann et al. 2007).
Arg
423
(subsites 1 and 1) in IS is part of the
highly conserved ‘RDP motif ’ in clan GH-J. Clearly,
it is part of a complex network of interactions in the
active site of FS enzymes (Meng & Fu
¨tterer 2003).
Changes in L. reuteri 121 IS residues at acceptor
subsite 1, R423K (also interacting with a glucosyl
residue at subsite 1; Arg
246
residue in B. subtilis
SacB), or W271N (Trp
85
residue in B. subtilis SacB),
caused altered FOS product patterns with sucrose,
yielding much less OSs and significantly more PSs
(Ozimek et al. 2006a). This suggests that modifica-
tion of amino acid residues, or the glycosidic binding
mode, at subsite 1 is important in determining the
product profile, i.e. PS vs. OS synthesis. Interest-
ingly, mutation R423H in L. reuteri IS virtually
abolished all enzymatic activity (Ozimek et al.
2006a). It was shown that the corresponding argi-
nine residue (R188) in exo-inulinase of family GH32
participates in substrate binding, is important for
recognition of the sugar ring and might be respon-
sible for specificity of the enzyme towards the
fructopyranosyl residue (Nagem et al. 2004).
The mutations W340N in L. reuteri IS (W163 in
B. subtilis LS SacB) and W172A in B. megaterium LS
greatly reduced enzyme activity and resulted in
synthesis of only small amounts of 1-kestose (IS)
and neokestose (LS; Homann et al. 2007; Ozimek
et al. 2006a). This residue does not directly interact
with the substrate, but forms the active-site bound-
ary at subsite 1. Removal of the large Trp side chain
lining the sucrose-binding pocket will enlarge the
size of subsite 1. Therefore, a tight fit of the
fructose moiety of sucrose in this subsite is clearly
favourable for the enzyme reaction.
Mutation W271N in L. reuteri IS (W85 in B.
subtilis LS SacB), an amino acid residue present in
many GH32 members, located at the bottom of the
active site near subsite 1, forming a hydrogen bond
with the C6 hydroxyl group of fructose) resulted in a
drastic drop in enzyme activity and in a three-fold
reduction in affinity for sucrose. Interestingly,
W271N synthesized larger fructan products com-
pared with the wild-type enzyme (Ozimek et al.
2006a).
B. megaterium LS residue Asn
252
(subsite 2)
clearly plays an important role in transfructosylation
(Homann et al. 2007). Substitution of Asn
252
to Ala
or Gly completely abolished PS production without
significantly affecting K
m
and k
cat
values, but causing
these variants to switch from mainly PS synthesis to
hydrolysis. In contrast, mutation N252D reduced
PS synthesis without changing K
m
and k
cat
values
significantly. The X-ray structure of the B. subtilis
SacB (74% amino acid identity with B. megaterium
LS) provides insights into the function of this Asn
252
residue. Its position at subsite 2 may allow it to
stabilise the third fructosyl unit of the growing OS
chain and direct it as an acceptor substrate into
the optimal position for further transfructosylation
(Homann et al. 2007).
The mutation N252D in B. megaterium LS clearly
reduced the coordination of a fructosyl unit at
subsite 2; however, it still occurred, as indicated
by the residual formation of PS. These findings
agree with a proposed model of the sugar-binding
subsites in two L. reuteri 121 FSs (Ozimek et al.
2006b). For fructan synthesis, subsites 2 and 3
need high binding affinity for the growing fructan
34 S. Kralj et al.

polymer chain. No structural determinants in FS
enzymes have been reported for additional acceptor
subsites (2, 3) so far.
Residues Arg
370
and Asn
252
, which appear to be
crucial for polyfructan synthesis, are conserved in
LSs from Gram-positive bacteria. In contrast, the
endophytic Gram-negative bacterium G. diazotrophi-
cus SRT4 secretes a constitutively expressed LS
(LsdA, EC 2.4.1.10), which mainly converts sucrose
into FOSs. It contains His
419
instead of Arg
370
at the
equivalent position (Martinez-Fleites et al. 2005).
This His residue is strictly conserved in Gram-
negative LSs. Furthermore, Asn
252
is strictly con-
served in LSs of Gram-positive bacteria, while
Gram-negative bacteria possess a conserved Arg
residue in the equivalent position (Homann et al.
2007). This section of the 3D structure of LsdA is
almost perfectly superimposable with the equivalent
residues of B. subtilis SacB in the active site. In
contrast the region constituting subsite 2 shows
clear differences (data not shown, Figure 1).
Expansion of substrate and product specificity
Although the substrate sucrose offers the advan-
tage of a high-energy glycosidic bond, similar to
that of nucleotide-activated sugars, it limits transfer
to glucose and fructose. The Gibbs energy change
for sucrose hydrolysis is DG
0
A
26.5 kJ mol
1
(298
K, pH 5.65; Goldberg & Tewari 1989; Goldberg
et al. 1989). This represents a high energy value
compared to other disaccharides, whose DG
0
A
values
for hydrolysis are 7, 8.8, and 15.5 kJ mol
1
for
isomaltose, lactose and maltose, respectively (Tewari
et al. 1991). This difference in the DG
0
A
value of
sucrose hydrolysis is available for synthesis of OSs
and PSs by sucrases using sucrose as a substrate (see
above). Those pathways can be expanded to sucrose
Figure 1. Local identification of amino acid residues crucial for structure-function relationships in FSs. Superimposition showing an
overlay of the 3D structures of the LSs from B. subtilis (Bs, blue; Meng & Fu
¨tterer 2003; accession code: 1PT2) and G. diazotrophicus (Gd,
grey; Martinez-Fleites et al. 2005; accession code: 1W18) with sucrose in the active site. Indicated are residues experimentally shown to be
important in structure-function relationships (FOS vs. PS synthesis) in FSs (Chambert & Petit-Glatron 1991; Homann et al. 2007; Ozimek
et al. 2006a). The analogous amino acid residues of B. megaterium (Bm) LS and L. reuteri IS are given in parentheses. (The figure was
generated with pyMOL 0.99rc6, DeLano Scientific LLC).
Fructansucrase enzymes and sucrose analogues 35

analogues, i.e. b-
D
-fructofuranosyl-a-
D
-galactopyra-
noside, b-
D
-fructofuranosyl-a-
D
-fucopyranoside and
b-
D
-fructofuranosyl-a-
D
-xylopyranoside (GalFru,
D
-Fuc-Fru, Xyl-Fru etc.; Baciu et al. 2005; Seibel
et al. 2005, 2006c). These may be used as new
substrates for OS and PS synthesis:
.using FSs in pathway A for the synthesis of new
glycopyranosyloligofructosides:
.alternatively in pathway B using GSs for glyco-
pyranoside transfer (Figure 2; Seibel et al.
2006b).
The transfer of either fructose or the glycopyrano-
side to other sugars, or to different natural products
as acceptors, is a challenging perspective. Advan-
tages of this sucrase system compared to the
currently employed enzymes, such as glycosynthases
and glycosyltransferases (Hancock et al. 2006;
Trincone et al. 2006), are that industrially estab-
lished GSs and FSs, and their genetically engi-
neered, improved, variants, may be usable for
extended substrate and product spectra.
Synthesis of sucrose analogues
The functionality of sucrose, with eight nearly
equivalent hydroxyl groups, makes selective chemi-
cal synthesis of derivatives laborious and difficult.
Sucrose analogues are unnatural activated substrates
that may serve for the synthesis of new OSs and PSs.
They provide a convenient and low-cost access to
synthesis and glycosidation. So far three different
approaches to sucrose analogues have been devel-
oped: chemical, enzymatic and chemo-enzymatic
synthesis (Figure 3). Enzymatic synthesis of the
sucrose analogues has been recently established for
a variety of structures. The exo-fructosyltransferase
(EC 2.4.1.162) from B. subtilis NCIMB 11871
transfers the fructosyl residue of sucrose to mono-
saccharide
D
-glucopyranosyl acceptors (
D
-mannose,
D
-galactose, 2-deoxy-
D
-glucose,
D
-fucose,
D
-xylose)
to yield the b-
D
-fructofuranosyl-a-
D
-glycopyrano-
sides (
D
-ManFru,
D
-GalFru,
D
-2-deoxy-Glc
Fru,
D
-FucFru,
D
-XylFru; Baciu et al. 2005;
Seibel et al. 2006c). A range of these are formed
in good or high yields, with respect to the
Figure 2. Enzymatic synthesis concept for OSs using sucrose analogues as substrates and FSs and GSs in pathways A and B. In pathway A,
unique fructans and FOSs are synthesized with a terminal glucose analogue (e.g. mannose, allose and xylose, etc). In pathway B, novel
glycan PSs and OSs may be synthesized, different from the normal glucan- and gluco-OSs, e.g. mannan- allan and xylan-type PSs and OSs.
(Seibel et al. 2006b).
36 S. Kralj et al.

quasi-equilibrium, under kinetic control and optimal
conditions. Use of l-glycopyranosides as acceptor
substrates results in formation of b-
D
-fructofurano-
syl-a-l-glycopyranoside (l-GlcFru, l-GalFru, l-
FucFru, l-XylFru, RhaFru), sucrose analogues
with b-1,2-glycosidic linkages (Seibel et al. 2006c).
Those b-1,2-glycosidic linkages are found in nature
as nucleotide-activated l-sugars, i.e. uridine dipho-
sphate (UDP)-a-l-arabinose and the GDP-a-l deri-
vative of fucose (Fuc) and cytidine monophosphate
activated sialic acid (b-CMP-NeuAc).
In contrast, the chemically equivalent non-enzy-
matic organic synthesis of
D
-GalFru occurs in
six synthetic steps (Seibel et al. 2005). The synthe-
tic approach additionally gives access to further
deoxyglycopyranosyl derivatives like 2-deoxy-2-
Figure 3. Different routes to sucrose analogues. A) Enzymatic using the levansucrase from B. subtilis. B) Chemo-enzymatic using the
dehydrogenase enzyme from A. tumefaciens followed by chemical modifications of the keto group. C) Chemical synthesis (2-deoxy-2-
aminoacetyl-D-Glc-Frc). The red circles show modifications in the various sucrose analogues compared to sucrose.
Fructansucrase enzymes and sucrose analogues 37

acetamido-b-
D
-fructofuranosyl-a-
D
-glucopyrano-
side (Lichtenthaler & Mondel 1997).
In the chemoenzymatic approach a range of new
sucrose derivatives can be obtained from 3-ketosu-
crose in aqueous medium with few reaction steps.
Agrobacterium tumefaciens produces a dehydrogenase
that oxidizes sucrose specifically at the 3-position,
yielding a-
D
-ribo-hex-3-ulopyranosyl-b-
D
-fructofur-
anoside (3-ketosucrose) (Figure 3). The biotrans-
formation was optimized and scaled up to give 70%
yield of 3-ketosucrose (Stoppok et al. 1992). Sub-
sequent chemical steps provide access to a range of
products, including a-
D
-allopyranosyl b-
D
-fructo-
furanoside (allosucrose; Timme et al. 1998), the
corresponding cyanohydrin (Timme et al. 2001),
oximes, and 3-amino-3-deoxy-a-
D
-allopyranosyl
a-
D
-fructofuranoside via the classical route of re-
ductive amination (Pietsch et al. 1994). Aminoacyl
and peptide conjugates were obtained through con-
ventional peptide synthesis with activated and pro-
tected amino acids (Figure 3). Deprotection yielded
new glyco-derivatives having unconventional substi-
tution patterns, namely 3-(aminoacylamino) allosac-
charides. Furthermore, Grignard reactions gave
CC linked derivatives with alkyl, as well as allyl
substituents (Anders et al. 2006).
Proof of concept: Sucrose analogues as novel
substrates for fructansucrase
Recently, it has been shown that both LSs and ISs
can use a wide range of sucrose analogues, except for
allosucrose, producing interesting new OSs and PSs
and demonstrating the first step towards a new route
for OS synthesis (Table I). The results of these
studies with
D
-ManFru,
D
-GalFru,
D
-XylFru,
and
D
-FucFru donors established that these su-
crose analogues are substrates for FSs. These
analogues have a glycosidic bond energy similar to
that of sucrose and thus should drive synthetic
reactions by non-Leloir glycosyl transferases.
D
-
ManFru,
D
-GalFru,
D
-XylFru, and
D
-FucFru
(all a-1,2-b-linked) were accepted by different LSs
like the LS of B. subtilis (Seibel et al. 2006a) and the
LS of L. reuteri (Biedendieck et al. 2007), thus
presenting proof of principle for this strategy. The
kinetic studies of the different substrates provided
deeper insights into the LS mechanism (Seibel et al.
2006c). So far from the kinetic and docking studies
of such sucrose analogues it can be assumed that the
FS enzyme possesses one donor/acceptor site that is
identical to the substrate site (Figure 4). Very
recently the preparative synthesis of 1-kestose and
1-nystose analogues (ManFru
2,3
, GalFru
2,3
, Xyl-
Fru
2,3
) from sucrose analogues has been acheived
with a b-fructofuranosidase from A. niger (Zucarro
et al. 2007).
Conclusion
Structure-function relationships in bacterial FSs are
not yet fully understood. However, structural fea-
tures determining the ratio of FOS versus polymer
produced are beginning to be revealed. Questions
that remain are how the type (i.e. in b-2,6 vs. b-2,1
glycosidic bond specificity) and size of OSs or PSs,
respectively, are determined.
New efficient methods now exist for enzymatic
synthesis of sucrose analogues. Most of these can be
used efficiently by FSs, several with high reaction
rates and good yields. Novel FOS synthesized with
this technology may be of potential use as prebiotics
or dietary fibre in food and health products, as well
as synthons to synthesize glycosylated drugs or
pharmaceutical ingredients.
However, a more extensive effort is required to
overcome several bottlenecks: low activity with some
of these sucrose analogues, and thus low reaction
rates and low reaction yields due to unfavourable
equilibria in some cases. To improve the potential of
this technology, the synthesis of these sucrose
analogues needs to be further optimized particularly
with respect to downstream processing, which may
represent a major bottleneck. Furthermore, the
Table I. Catalytic reactions of different fructansucrase with various sucrose analogues (D and L pyranoses)
1
.
Sucrose
D
-ManFru
D
-AllFru
D
-GalFru
D
-XylFru
D
-FucFru
LS L. mes. PS n.t. n.t. PS, OS PS, OS n.t.
IS L. citreum PS n.t n.t. PS PS, OS n.t.
LS B. subtilis PS P, OS H
1
PS,OS PS, OS PS, OS
LS B. meg. PS,OS,H n.t. n.t. H, PS PS, OS n.t.
LS L. reuteri PS, OS OS n.t. OS,PS PS, OS PS, OS, H
FF A. niger OS OS n.t. OS OS n.t.
1
Allosucrose is a very poor substrate compared with sucrose.
LS, levansucrase; IS, inulosucrase; FF, b-fructofuranosidase; L. mes., Leuconostoc mesenteroides;L. citreum,Leuconostoc citreum;B. subtilis,
Bacillus subtilis;B. meg., Bacillus megaterium;L. reuteri, Lactobacillus reuteri;A. niger, Aspergillus niger; n.t., not tested; H, hydrolysis; OS,
oligosaccharide formation; PS, polysaccharide formation.
When a specific reaction is dominant it is indicated in bold face.
38 S. Kralj et al.

different sucrases need to be optimized by rational
design/random mutagenesis for these novel sub-
strates. Another exciting possibility may be to use
these analogues in combination with GS enzymes to
synthesize novel galactan, mannan, and allan PSs
and OSs.
In addition, new FOSs may be synthesized by
using sucrose analogues where, instead of glucose,
the fructose moiety of sucrose is modified. The
relatively recent elucidation of LS 3D information,
together with the synthesis of sucrose analogues,
suggests a promising future for the development of
unique tailor-made FOSs.
In the future, novel FSs may be discovered that
are able to use the other hydroxyl moieties present
on the fructose molecule (as is the case for GSs) so
that besides the known b-2,6 and b-2,1 linkage
types, other types of linkages and thus other
saccharides may be synthesized.
Acknowledgements
Financial support by the German Research Founda-
tion via Sonderforschungsbereich 578 ‘From Gene
to Product’ and by the EET programme of the
Dutch government (project number EETK01129) is
gratefully acknowledged.
References
Alberto F, Bignon C, Sulzenbacher G, Henrissat B, Czjzek M.
2004. The three-dimensional structure of invertase (b-fructo-
sidase) from Thermotoga maritima reveals a bimodular arrange-
ment and an evolutionary relationship between retaining and
inverting glycosidases. J Biol Chem 279:1890318910.
Alberto F, Jordi E, Henrissat B, Czjzek M. 2006. Crystal structure
of inactivated Thermotoga maritima invertase in complex with
the trisaccharide substrate raffinose. Biochem J 395:457462.
Anders J, Buczys R, Lampe E, Walter M, Yaacoub E, Buchholz K.
2006. New regioselective derivatives of sucrose with amino acid
and acrylic groups. Carbohydr Res 341:322331.
Baciu IE, Jo
¨rdening HJ, Seibel J, Buchholz K. 2005. Investiga-
tions of the transfructosylation reaction by fructosyltransferase
from B. subtilis NCIMB 11871 for the synthesis of the sucrose
analogue galactosyl-fructoside. J Biotechnol 116:347357.
Baird JK, Longyear VM, Ellwood DC. 1973. Water insoluble and
soluble glucans produced by extracellular glycosyltransferases
from Streptococcus mutans. Microbios 8:143150.
Biedendieck R, Beine R, Gamer M, Jordan E, Buchholz K, Seibel
J, Dijkhuizen L, Malten M, Jahn D. 2007. Export, purification,
and activities of affinity tagged Lactobacillus reuteri levansucrase
produced by Bacillus megaterium. Appl Microbiol Biotechnol
74:10621073.
Binder TP, Coˆte´ GL, Robyt JF. 1983. Disproportionation
reactions catalyzed by Leuconostoc and Streptococcus glucansu-
crases. Carbohydr Res 124:275286.
Buchholz K, Monsan P. 2003. Dextransucrase. In: Whitaker JR,
Voragen AGJ, Wong DWS, editors. Handbook of food enzymol-
ogy. New York: Dekker. p 589603.
CAZy 2006. Carbohydrate-active enzymes. Available at: http://
www.cazy.org. (accessed 21 October 2006).
Cerning J. 1990. Exocellular polysaccharides produced by lactic
acid bacteria. FEMS Microbiol Rev 7:113130.
Figure 4. (A) Lowest-energy dockings of the substrates sucrose (left) and (B)
D
-GalFru (right) with levansucrase (LS) show identical
orientation in the enzyme active site. Further conformations of
D
-GalFru docking experiments are also superimposed (grey). Essential are
the distances of Asp
86
(attacking C-1 as nucleophile) and Glu
342
(protonating the glycosidic bond). (The Figure was generated with the
program AutoDock).
Fructansucrase enzymes and sucrose analogues 39

Chambert R, Gonzy-Treboul G. 1976. Levansucrase of Bacillus
subtilis. Characterization of a stabilized fructosyl-enzyme com-
plex and identification of an aspartyl residue as the binding site
of the fructosyl group. Eur J Biochem 71:493508.
Chambert R, Petit-Glatron MF. 1991. Polymerase and hydrolase
activities of Bacillus subtilis levansucrase can be separately
modulated by site-directed mutagenesis. Biochem J 279:3541.
Chambert R, Treboule G, Dedonder R. 1974. Kinetic studies of
levansucrase of Bacillus subtilis. Eur J Biochem 41:285300.
Coˆ te´ GL, Robyt JF. 1982. Isolation and partial characterization of
an extracellular glucansucrase from Leuconostoc mesenteroides
NRRL B-1355 that synthesizes an alternating (1 06), (103)-
a-
D
-glucan. Carbohydr Res 101:5774.
Davies GJ, Wilson KS, Henrissat B. 1997. Nomenclature for
sugar-binding subsites in glycosyl hydrolases. Biochem J
321:557559.
Doelle HW, Kirk L, Crittenden R, Toh H, Doelle MB. 1993.
Zymomonas mobilis Science and industrial application. Crit
Rev Biotechnol 13:5798.
Goldberg R, Tewari J. 1989. Thermodynamic and transport
properties of carbohydrates and their monophosphates: The
pentoses and hexoses. J Phys Chem Ref Data 18:809822.
Goldberg R, Tewari J, Ahluwalia J. 1989. Thermodynamics of the
hydrolysis of sucrose. J Biol Chem 264:99019904.
Gross M, Geier G, Rudolph K, Geider K. 1992. Levan and
levansucrase synthesized by the fireblight pathogen Erwinia
amylovora. Physiol Mol Plant Pathol 40:371381.
Hancock SM, Vaughan MD, Withers SG. 2006. Engineering of
glycosidases and glycosyltransferases. Curr Opin Chem Biol
10:509519.
Hehre EJ. 1941. Production from sucrose of a serologically
reactive polysaccharide by a sterile bacterial extract. Science
93:237238.
Hernandez L, Arrieta J, Menendez C, Vazquez R, Coego A,
Suarez V, Selman G, Petit-Glatron MF, Chambert R. 1995.
Isolation and enzymic properties of levansucrase secreted by
Acetobacter diazotrophicus SRT4, a bacterium associated with
sugar cane. Biochem J 309:113118.
Homann A, Biedendieck R, Gotze S, Jahn D, Seibel J. 2007.
Insights into polymer versus oligosaccharide synthesis: muta-
genesis and mechanistic studies of a novel levansucrase from
Bacillus megaterium. Biochem J 407:189198.
Kaur N, Gupka AK. 2002. Applications of oligofructose in health
and nutrition. J. Biosci 27:703714.
Korakli M, Rossmann A, Ganzle MG, Vogel RF. 2001. Sucrose
metabolism and exopolysaccharide production in wheat and
rye sourdoughs by Lactobacillus sanfranciscensis. J Agric Food
Chem 49:51945200.
Korakli M, Pavlovic M, Ganzle MG, Vogel RF. 2003. Exopoly-
saccharide and kestose production by Lactobacillus sanfrancis-
censis LTH2590. Appl Environ Microbiol 69:20732079.
Kralj S, van Geel-Schutten GH, Rahaoui H, Leer RJ, Faber EJ,
van der Maarel MJ, Dijkhuizen L. 2002. Molecular character-
ization of a novel glucosyltransferase from Lactobacillus reuteri
strain 121 synthesizing a unique, highly branched glucan with
a-(104) and a-(106) glucosidic bonds. Appl Environ Micro-
biol 68:42834291.
Kralj S, van Geel-Schutten GH, van der Maarel MJ, Dijkhuizen
L. 2004. Biochemical and molecular characterization of
Lactobacillus reuteri 121 reuteransucrase. Microbiology
150:20992112.
Lichtenthaler F, Mondel S. 1997. Manno- and altro-sucrose and
some amino-analogs. Carbohydr Res 303:293302.
Martinez-Fleites C, Ortiz-Lombardia M, Pons T, Tarbouriech N,
Taylor EJ, Arrieta JG, Hernandez L, Davies GJ. 2005. Cr ystal
structure of levansucrase from the Gram-negative bacterium
Gluconacetobacter diazotrophicus. Biochem J 390:1927.
Meng G, Fu
¨tterer K. 2003. Structural framework of fructosyl
transfer in Bacillus subtilis levansucrase. Nat Struct Biol 10:935
941.
Monchois V, Remaud-Simeon M, Russell RR, Monsan P, Will-
emot RM. 1997. Characterization of Leuconostoc mesenteroides
NRRL B-512F dextransucrase (DSRS) and identification of
amino-acid residues playing a key role in enzyme activity. Appl
Microbiol Biotechnol 48:465472.
Monchois V, Willemot RM, Monsan P. 1999. Glucansucrases:
Mechanism of action and structure-function relationships.
FEMS Microbiol Rev 23:131151.
Morales-Arrieta S, Rodriguez ME, Segovia L, Lopez-Munguia A,
Olvera-Carranza C. 2006. Identification and functional char-
acterization of levS, a gene encoding for a levansucrase from
Leuconostoc mesenteroides NRRL B-512 F. Gene 376:5967.
Nagem RA, Rojas AL, Golubev AM, Korneeva OS, Eneyskaya
EV, Kulminskaya AA, Neustroev KN, Polikarpov I. 2004.
Crystal structure of exo-inulinase from Aspergillus awamori:
The enzyme fold and structural determinants of substrate
recognition. J Mol Biol 344:471480.
Nakakuki T. 2002. Present status and future of functional
oligosaccharide development in Japan. Pure Appl Chem
74:12451251.
Olivares-Illana V, Wacher-Odarte C, Le Borgne S, Lopez-Mun-
guia A. 2002. Characterization of a cell-associated inulosucrase
from a novel source: a Leuconostoc citreum strain isolated from
Pozol, a fermented corn beverage of Mayan origin. J Ind
Microbiol Biotechnol 28:112117.
Ozimek LK, van Hijum SA, van Koningsveld GA, van Der Maarel
MJ, van Geel-Schutten GH, Dijkhuizen L. 2004. Site-directed
mutagenesis study of the three catalytic residues of the
fructosyltransferases of Lactobacillus reuteri 121. FEBS Lett
560:131133.
Ozimek LK, Kralj S, Kaper T, van der Maarel MJ, Dijkhuizen L.
2006a. Single amino acid residue changes in subsite 1 of
inulosucrase from Lactobacillus reuteri 121 strongly influence
the size of products synthesized. FEBS J 273:41044113.
Ozimek LK, Kralj S, van der Maarel MJ, Dijkhuizen L. 2006b.
The levansucrase and inulosucrase enzymes of Lactobacillus
reuteri 121 catalyse processive and non-processive transglyco-
sylation reactions. Microbiology 152:11871196.
Pietsch M, Walter W, Buchholz K. 1994. Regioselektive synthesis
of new sucrose derivatives via 3-ketosucrose. Carbohydr Res
254:183194.
Rastall RA, Hotchkiss AT. 2003. Potential for the development of
prebiotic oligosaccharides from biomass. ACS Symp Ser 2003.
Washington, DC: American Chemical Society.
Remaud-Simeon M, Lopez-Munguia A, Pelenc V, Paul F,
Monsan P. 1994. Production and use of glucosyltransferases
from Leuconostoc mesenteroides NRRL B-1299 for the synthesis
of oligosaccharides containing alpha-(102) linkages. Appl
Biochem Biotechnol 44:101117.
Rosell KG, Birkhed D. 1974. An inulin-like fructan produced by
Streptococcus mutans, strain JC2. Acta Chem Scand B 28:589.
Seibel J, Moraru R, Go
¨tze S. 2005. Biocatalytic and chemical
investigations in the synthesis of sucrose analogues. Tetrahe-
dron 61:70817086.
Seibel J, Beine R, Moraru R, C B, Buchholz K. 2006a. A new
pathway for the synthesis of oligosaccharides by the use of non-
Leloir glycosyltransferases. Biocatal Biotrans 24:157165.
Seibel J, Buchholz K, Jo
¨rdening HJ. 2006b. Glycosylation with
activated sugars using glycosyltransferases and transglycosi-
dases. Biocatal Biotrans 24:311342.
Seibel J, Moraru R, Gotze S, Buchholz K, Na’amnieh S,
Pawlowski A, Hecht HJ. 2006c. Synthesis of sucrose analogues
and the mechanism of action of Bacillus subtilis fructosyltrans-
ferase (levansucrase). Carbohydr Res 341:23352349.
40 S. Kralj et al.

Skov LK, Mirza O, Sprogoe D, Dar I, Remaud-Simeon M,
Albenne C, Monsan P, Gajhede M. 2002. Oligosaccharide and
sucrose complexes of amylosucrase. Structural implications for
the polymerase activity. J Biol Chem 277:4774147747.
Song KB, Seo JW, Rhee SK. 1999. Transcriptional analysis of
levU operon encoding saccharolytic enzymes and two apparent
genes involved in amino acid biosynthesis in Zymomonas mobilis.
Gene 232:107114.
Steinmetz M, Le Coq D, Aymerich S, Gonzy-Treboul G, Gay P.
1985. The DNA sequence of the gene for the secreted Bacillus
subtilis enzyme levansucrase and its genetic control sites. Mol
Gen Genet 200:220228.
Stoppok E, Matalla K, Buchholz K. 1992. Microbial modification
of sugars as building blocks for chemicals. Appl Microb
Biotechnol 36:604610.
Tewari J, Goldberg R. 1991. Thermodynamics of hydrolysis of
disaccharides. Biophys Chem 40:5967.
Timme V, Buczys R, Buchholz K. 1998. Kinetic investigations on
the hydrogenation of 3-ketosucrose. Starch/Sta
¨rke 50:2932.
Timme V, Buczys R, Buchholz K. 2001. Cyanhydrin-Derivat der
Saccharose via 3-Ketosaccharose. Chem Ing Tech 73:1357
1361.
Trincone A, Giordano A. 2006. Glycosyl hydrolases and glycosyl-
transferases in the synthesis of oligosaccharides. Curr Org
Chem 10:11631193.
van Hijum SA, Bonting K, van der Maarel MJ, Dijkhuizen L.
2001. Purification of a novel fructosyltransferase from Lacto-
bacillus reuteri strain 121 and characterization of the levan
produced. FEMS Microbiol Lett 205:323328.
van Hijum SA, van Geel-Schutten GH, Rahaoui H, van der
Maarel MJ, Dijkhuizen L. 2002. Characterization of a novel
fructosyltransferase from Lactobacillus reuteri that synthesizes
high-molecular-weight inulin and inulin oligosaccharides. Appl
Environ Microbiol 68:43904398.
van Hijum SA, Szalowska E, van der Maarel MJ, Dijkhuizen L.
2004. Biochemical and molecular characterization of a levan-
sucrase from Lactobacillus reuteri. Microbiology 150:621630.
van Hijum SA, Kralj S, Ozimek LK, Dijkhuizen L, van Geel-
Schutten IG. 2006. Structure-function relationships of glucan-
sucrase and fructansucrase enzymes from lactic acid bacteria.
Microbiol Mol Biol Rev 70:157176.
Verhaest M, Ende WV, Roy KL, De Ranter CJ, Laere AV, Rabijns
A. 2005. X-ray diffraction structure of a plant glycosyl
hydrolase family 32 protein: Fructan 1-exohydrolase IIa of
Cichorium intybus. Plant J 41:400411.
Werner RG, Kopp K, Schlueter M. 2007. Glycosylation of
therapeutic proteins in different production systems. Acta
Paediatr Suppl 96:1722.
Wong CH. 2005. Protein glycosylation: New challenges and
opportunities. J Org Chem 70:42194225.
Yanase H, Maeda M, Hagiwara E, Yagi H, Taniguchi K, Okamoto
K. 2002. Identification of functionally important amino acid
residues in Zymomonas mobilis levansucrase. J Biochem (Tokyo)
132:565572.
Yun JW. 1996. Fructooligosaccharides Occurence, preparation
and application. Enz Microb Technol 19:107117.
Zuccaro A, Gotze S, Kneip S, Dersch P, Seibel J. 2007. Taylor
made fructooligosaccharides by combination of substrate and
genetic engineering. ChemBioChem. (in press).
Fructansucrase enzymes and sucrose analogues 41