Probing substrate promiscuity of amylosucrase from neisseria polysaccharea

Article abstract of DOI:10.1002/cctc.201300012

The amylosucrase from Neisseria polysaccharea (NpAS) naturally catalyzes the synthesis of a variety of products from sucrose and shows signs of plasticity of its active site. To explore further this promiscuity, the tolerance of amylosucrase towards different donor and acceptor substrates was investigated. The selection of alternate donor substrates was first made on the basis of preliminary molecular modeling studies. From 11 potential donors harboring selective derivatizations that were experimentally evaluated, only p-nitrophenyl-α-D-glucopyranoside was used by the wild-type enzyme, and this underlines the high specificity of the -1 subsite of NpAS for glucosyl donor substrates. The acceptor substrate promiscuity was further explored by screening 20 hydroxylated molecules, including D- and L-monosaccharides as well as polyols. With the exception of one compound, all were successfully glucosylated, and this showcases the tremendous plasticity of the +1 subsite of NpAS, which is responsible for acceptor recognition. The products obtained from the transglucosylation reactions of three selected acceptors were characterized, and they revealed original structures and enzyme enantiopreference, which were more particularly analyzed by insilico docking analyses.

Full text of DOI:10.1002/cctc.201300012

CHEMCATCHEM
FULL PAPERS
DOI: 10.1002/cctc.201300012
Probing Substrate Promiscuity of Amylosucrase from
Neisseria polysaccharea
David Daudꢀ, Elise Champion, Sandrine Morel, David Guieysse, Magali Remaud-Simꢀon, and
Isabelle Andrꢀ*[a]
The amylosucrase from Neisseria polysaccharea (NpAS) naturally
catalyzes the synthesis of a variety of products from sucrose
and shows signs of plasticity of its active site. To explore fur-
ther this promiscuity, the tolerance of amylosucrase towards
different donor and acceptor substrates was investigated. The
selection of alternate donor substrates was first made on the
basis of preliminary molecular modeling studies. From 11 po-
tential donors harboring selective derivatizations that were ex-
perimentally evaluated, only p-nitrophenyl-a-d-glucopyrano-
side was used by the wild-type enzyme, and this underlines
the high specificity of the ꢀ1 subsite of NpAS for glucosyl
donor substrates. The acceptor substrate promiscuity was fur-
ther explored by screening 20 hydroxylated molecules, includ-
ing d- and l-monosaccharides as well as polyols. With the ex-
ception of one compound, all were successfully glucosylated,
and this showcases the tremendous plasticity of the +1 sub-
site of NpAS, which is responsible for acceptor recognition. The
products obtained from the transglucosylation reactions of
three selected acceptors were characterized, and they revealed
original structures and enzyme enantiopreference, which were
more particularly analyzed by in silico docking analyses.
Introduction
Carbohydrates are involved in many biological processes, and
they are known to be common building blocks of life. Storage
reserve, nutrition, genetic and metabolic pathways, and cell
membranes commonly involve carbohydrates and glycoconju-
gates. Given the wide range of biological roles of carbohy-
drates, it is not surprising to find a broad variety of glycostruc-
tures in nature. The structural diversity of carbohydrates essen-
tially emanates from the complexity of the constituting alpha-
bet of glucosyl units and the types of possible osidic linkages.
In this context, enzymes are powerful tools to access original
structures. Their exquisite specificity helps to circumvent the
tedious problem of protection/deprotection of highly reactive
hydroxyl groups encountered in the chemical syntheses of
sugars. In addition, the potential offered by their natural or en-
gineered substrate promiscuity is of utmost interest to synthe-
size novel carbohydrate-based molecules.[1,2] In particular, sub-
strate promiscuity of glycosyltransferases towards activated nu-
cleotide sugars has already been exploited to glycosylate origi-
nal carbohydrate structures with various osidic patterns.[3,4]
However, the need for activated sugars as glycosyl donors im-
pairs the utilization of such enzymes, as these sugars are rela-
tively scarce. Therefore, the search for alternative sugar biosyn-
thetic enzymes that use readily available glycosyl donor sub-
strates and that are able to act on non-natural substrates pro-
vides an opportunity to fill the reservoir of accessible carbohy-
drate structures. Within this context, transglycosylases such as
glucansucrases, which use widespread sucrose (a-d-glucopyra-
nosyl-1,2-b-d-fructofuranoside) as a donor substrate, are of par-
ticular interest. Among this class of enzymes, amylosucrases
are members of the glycoside–hydrolase[5] (GH) family 13 ac-
cording to carbohydrate-active enzymes (CAZy) classification.[6]
They are produced by various strains and have been character-
ized from Alteromonas, Arthrobacter, Deinococcus, and Neisseria
species.[7] They naturally catalyze the synthesis of a-1,4-glu-
cans, together with some secondary reactions (sucrose isomeri-
zation and hydrolysis). Interestingly, amylosucrases have also
been shown to glucosylate various non-natural hydroxylated
acceptors such as flavonoids, hydroquinone, catechin, and car-
bohydrates.[8–12] Although some molecules have already been
described as amylosucrase acceptors, only a few donors have
been reported up to now. Besides sucrose, which is known to
be the natural donor substrate of amylosucrases, fluorogluco-
sides and maltooligosaccharides can also act as glucosyl
donors.[13,14]
[a] D. Daudꢀ, Dr. E. Champion, Dr. S. Morel, Dr. D. Guieysse,
Prof. M. Remaud-Simꢀon, Dr. I. Andrꢀ
Additionally, two sucrose analogues have been reported as
possible substrates for transglycosylation reactions. a-d-Galac-
topyranosyl-1,2-b-d-fructofuranoside (GalFru) and allosucrose
have indeed been described as potential substrates of Neisseria
perflava amylosucrase and an unspecified histidine-tagged
amylosucrase, respectively, although no fine characterization of
the reaction products was carried out.[15,16] However, in a prior
report, allosucrose was not recognized by the amylosucrase
Universitꢀ de Toulouse, INSA,UPS INP; LISBP, 135 Avenue de Rangueil,
F-31077 Toulouse, France
CNRS, UMR5504, F-31400 Toulouse, France
INRA, UMR792 Ingꢀnierie des Systꢁmes Biologiques et des Procꢀdꢀs,
F-31400 Toulouse, France
Fax: (+33)561559400
E-mail: isabelle.andre@insa-toulouse.fr
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cctc.201300126.
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 2288 – 2295 2288

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
from N. perflava, and 3-deoxysucrose, 6-deoxysucrose, 6-deoxy-
fluorosucrose, and 4,6-dideoxysucrose were also not recog-
nized.[17] Nevertheless, sucrose derivatives could constitute a re-
newable attractive biosource for glycodiversification and in-
crease the number of accessible glycoconjugates through the
catalysis of promiscuous sucrose-utilizing enzymes, as first pro-
posed by Buchholz and co-workers who also investigated the
promiscuity of GH family 70 enzymes towards alternate
acceptors.[18–23]
sive investigation of the substrate promiscuity of NpAS. Both
experimental and theoretical results are discussed with regard
to the activity of NpAS towards its natural substrate, sucrose.
Altogether, the collected data contribute to the better under-
standing of the inter-relationships between structure and activ-
ity of this class of enzymes, and hopefully, the results will help
to promote the utilization of NpAS in biotechnological
applications.
Until recently, the 3D structure of Neisseria polysaccharea
amylosucrase (NpAS) was the only available structure of an
amylosucrase.[24–27] With the determination of the X-ray crystal-
lographic structure of the thermostable amylosucrase from De-
inococcus geothermalis,[28] there has been a new wealth of
structural information to better understand substrate and
product specificities of amylosucrase. Indeed, both crystallo-
graphic structures revealed a deep active site pocket in which
sucrose binds to the bottom and occupies two subsites num-
bered ꢀ1 and +1 according to GH nomenclature (Fig-
ure 1a).[29] The ꢀ1 subsite is involved in the specific recogni-
tion of the glucosyl moiety of sucrose. The residues from the
first coordination sphere constituting this subsite are repre-
sented in Figure 1b and have been identified to play a role in
catalysis, stacking, and stabilization during the hydrolysis of
the osidic bond in sucrose and the formation of a covalent b-
d-glucosyl–enzyme.[30,31] A dense hydrogen-bonding network is
responsible for sucrose recognition and is highly specific for
the glucosyl unit, as the four hydroxyl substituents are in-
volved in interactions with active-site residues. The +1 subsite
has a dual role, as it is implicated both in the recognition of
the fructosyl moiety of sucrose and in acceptor binding. Fur-
thermore, it was shown to be tolerant to mutations performed
with the aim of modifying the recognition of non-natural ac-
ceptor substrates. To gain further insight into the active site
plasticity of NpAS, substrate derivatives were evaluated as po-
tential donors and acceptors. We report herein the first exten-
Results and Discussion
To decipher the substrate promiscuity of NpAS, a panel of 11
potential donors and 20 acceptors was tested experimentally.
As a result of the stringent specificity of this enzyme for su-
crose as a donor substrate, only sucrose derivatives with slight
modifications to the sucrose scaffold were evaluated. These
substrates display broad diversity in terms of both substituents
and substitution positions on either the glucosyl or the fructo-
syl moiety. Regarding the acceptor substrates, the selected
compounds included d- and l-monosaccharides as well as
sugar alcohols
Evaluation of sucrose analogues as potential donors
Ester, keto, epimer, phosphate, glycosyl, and chloro derivatives
of sucrose were examined. The targeted molecules are repre-
sented in Figure 2. Five of them were enzymatically synthe-
sized from sucrose (1) to give 3-ketosucrose (2), sucrose-6-ace-
tate (S6A, 3), sucrose-6,6’-diacetate (S66A, 4), a-d-galactopyra-
nosyl-1,2-b-d-fructofuranoside (GalFru, 5), and a-d-xylopyrano-
syl-1,2-b-d-fructofuranoside (XylFru, 6), and the six others were
commercially available, namely, glucose-1-phosphate (G1P, 7),
p-nitrophenyl-a-d-glucopyranoside (pNP-Glc, 8), p-nitrophenyl-
a-d-galactopyranoside (pNP-Gal, 9), p-nitrophenyl-a-d-manno-
pyranoside (pNP-Man, 10), a-d-galactopyranosyl-1,4-a-d-gluco-
pyranoside-1,2-b-d-fructofuranoside (raffinose, 11), and 1,6-di-
Figure 1. Representation of sucrose bound in the active site of NpAS. The structure of wild-type amylosucrase from Neisseria polysaccharea as a complex with
the sucrose donor (PDB id: 1JGI) was 4 ꢂ sculpted (a) and the residues involved in substrate recognition are represented in 3D (b) and 2D projections (c).
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 2288 – 2295 2289

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
energy values are equivalent to
that of sucrose, and this indi-
cates clearly that such calcula-
tions only hint towards the geo-
metric and electrostatic compati-
bility of the molecule with the
active site of the enzyme but
may not be correlated to sub-
strate reactivity. The sucrose de-
rivatives were then assayed as
glycosyl donors. None of the
tested molecules were used as
donors, except pNP-Glc. This
confirms the high specificity of
the ꢀ1 subsite of NpAS for the
glucosyl moiety. The network of
hydrogen bonds and p-stacking
interactions occurring around
the glucosyl moiety of the natu-
ral donor is too dense to tolerate
any of the substituents intro-
duced onto the glucosyl ring
(Figure 1c). Of note, the alterna-
tive pNP-Glc donor was reported
to be a substrate of glucansu-
crases from the GH70 family.[32]
The kinetic parameters of the
reaction catalyzed by pNP-Glc
(8) were determined by measur-
Figure 2. Sucrose analogues evaluated as potential donor substrates of NpAS: sucrose (1), 3-ketosucrose (2), su-
crose-6-acetate (3), sucrose-6,6’-acetate (4), a-d-galactopyranosyl-1,2-b-d-fructofuranoside (5), a-d-xylopyranosyl-
1,2-b-d-fructofuranoside (6), glucose-1-phosphate (7), p-nitrophenyl-a-d-glucopyranoside (8), p-nitrophenyl-a-d-
galactoside (9), p-nitrophenyl-a-d-mannoside (10), a-d-galactopyranosyl-1,4-a-d-glucopyranoside-1,2-b-d-fructo-
furanoside (11), and 1,6-dichloro-1,6-dideoxy-b-d-fructofuranosyl-4-chloro-4-deoxy-a-d-galactopyranoside (12).
chloro-1,6-dideoxy-b-d-fructofuranosyl-4-chloro-4-deoxy-a-d-
galactopyranoside (sucralose, 12).
ing the absorbance at 405 nm of released p-nitrophenoxide
(see the Supporting Information, Figure S1a,b). The kinetic
data is well fitted by a Michaelis–Menten model (Figure S1c).
The Michaelis constant value, KM, towards pNP-Glc was found
to be higher than that observed for sucrose, whereas the maxi-
mum velocity and the catalytic efficiency values (kcat/KM) were
greatly reduced (Table 1). Interestingly, the formation of p-ni-
trophenylglucooligosaccharides was detected by mass spec-
A preliminary molecular modeling study was first carried out
to evaluate the geometrical and electrostatic compatibility of
the sucrose derivatives with the active site of NpAS. Docking
of each analogue in the ꢀ1 and +1 subsites of NpAS was
compared to the docking of sucrose into these subsites. As
shown in Figure 3, van der Waals electrostatic contributions
and total energies reveal that all of the sucrose analogues are
geometrically compatible with the active site of NpAS, al-
though rearrangement of some amino acid side chains had to
occur to accommodate bulkier substituents. Indeed, docking
of raffinose (11) entailed the reorientation of the Asp144,
Tyr147, and His187 amino acid side chains to enable accom-
modation of the trisaccharide galactosyl moiety. Docking of su-
crose esters S6A (3) and S66A (4) induced a 908 rotation of the
His187 residue to allow binding of the acetate substituent. In
all the minimized structures, the three catalytic residues
(Asp286, Glu328, and Asp393) adopted a conformation compa-
rable to that found for the crystallographic structure of the
NpAS–sucrose complex [protein database (PDB) code: 1JGI],
and this indicates that catalysis could occur (Figure 1). Interest-
ingly, for each analogue, the values of the interaction energy
were lower than that calculated for 1-fluoro-1-deoxy-a-d-glu-
cose, a molecule previously reported to act as a donor sub-
strate of NpAS.[14] Sucrose derivatives described in the literature
as inhibitors or nondonor substrates of NpAS[17] were also in-
troduced in these calculations. As shown in Figure 3, the
Figure 3. Energies of interaction of NpAS complexed with sucrose deriva-
tives. Black, unshaded, and shaded bars represent van der Waals, electrostat-
ic contributions, and total energies, respectively. The reference sucrose is
highlighted in grey, known substrates of amylosucrases in green, known in-
hibitors or nonsubstrate molecules of amylosucrases in red, and nonassayed
sucrose derivatives in blue.
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 2288 – 2295 2290

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
donor, and the catalytic efficiency value towards
pNP-Glc was much lower than that determined for
sucrose. As a consequence, these results emphasize
the strong stringency of sucrose recognition by
NpAS.
Table 1. Kinetic parameters of NpAS towards pNP-Glc compared to those of NpAS to-
wards sucrose.
KM
[mm]
kcat
[minꢀ1
Vmax
kcat/KM
]
[mmolproduct released minꢀ1 gꢀ1] [mmꢀ1 minꢀ1
]
[Sucrose][a] <20 mm 1.7
>20 mm 38.7
[pNP-Glc] <30 mm 66.9ꢁ10.2 1.76ꢁ0.05 22.3ꢁ2.6
20
33
288
472
11.76
0.85
0.03
The active site was designed through natural evo-
lution to accommodate sucrose. However, informa-
tion inferred from the unrecognized substrates helps
to understand structure–activity relationships and to
guide molecular design experiments to extend the
scope of reactions catalyzed by NpAS. In particular,
[a] Data for sucrose taken from Ref. [33].
trometry, which shows that pNP-Glc (8) is utilized by NpAS as
both a donor and an acceptor substrate. The other tested p-ni-
trophenylglycosides, namely, pNP-Man (10) and pNP-Gal (9),
which are C-2 and C-4 epimers of pNP-Glc (8), respectively,
were not substrates of NpAS, and this emphasizes the impor-
tance of the orientation of the hydroxyl group at C-4 and C-2
in the reactivity of sucrose. Indeed, the axial orientation of the
C-4 hydroxyl group probably induces a loss of stabilizing hy-
drogen-bonding interactions with Asp144 and Arg509 and
stacking interactions with Tyr147. In pNP-Gal (9) as well as in
GalFru (5), this may contribute to a loss of reactivity. In the
case of sucralose (12), the chlorine at the C-4 axial position
also prevents the formation of stabilizing hydrogen bonds at
this position. Allosucrose has been reported as a donor sub-
strate of amylosucrase.[16] This suggests a possible tolerance to
modifications at the C-3 position. However, in our tests, 3-keto-
sucrose (2) was not used by the enzyme. This may be attribut-
ed to the disruption of hydrogen bonds between the hydroxyl
substituent and the His392 and Asp393 residues. The involve-
ment of the hydroxyl group at C-6 in sucrose reactivity was
also highlighted by using S6A (3), S66A (4), and less bulky
XylFru (6) as alternative donors. In these molecules, modifica-
tion or the absence of the C-6 hydroxyl group could presuma-
bly prevent stabilization by both the Asp286 and His187 resi-
dues. Moreover, the leaving group that binds at the +1 subsite
of NpAS also plays a key role in substrate reactivity. Indeed,
glucose-1-phosphate could not be used by NpAS as a glucosyl
the promiscuity towards pNP-Glc, herein revealed, might be
particularly attractive for the development of colorimetric-
based screening assays and could be envisioned for easy high-
throughput detection of amylosucrase activity. Such screening
methods could be particularly efficient to isolate NpAS mutants
active onto other pNP-glycosyl derivatives, which are easier to
modify than sucrose-based molecules.
Evaluation of acceptor candidates for glucosylation by NpAS
NpAS naturally catalyzes the hydrolysis of the disaccharide su-
crose, and released glucose is used as an acceptor for the syn-
thesis of a-glucans. To investigate the potential of NpAS to uti-
lize non-natural acceptors, we tested a library of glucose ana-
logues. The selected acceptors included both l- and d-mono-
saccharides (arabinose, galactose, altrose, fucose, xylose, allose,
and mannose) and sugar alcohols (d-sorbitol, xylitol, d-arabitol,
d-mannitol, myoinositol, and maltitol) by using sucrose as the
glucosyl donor. As shown in Figure 4, 19 out of the 20 accept-
ors assayed were recognized by NpAS, and the conversion
reached 100% in the case of l-altrose. These results highlight
the plasticity of the +1 subsite, which is responsible for ac-
ceptor binding, and also show the broad natural promiscuity
towards alditols, as all six selected molecules were successfully
glucosylated with conversion rates ranging from 25% for myoi-
nositol up to 84% for xylitol. Of note, the experiments carried
out with l- and d-monosaccharides clearly underline the enan-
Figure 4. Transglucosylation of various monosaccharides in the d (blue) and l (red) series and alditols (green) by using NpAS at an acceptor/sucrose molar
ratio of 2:1 (292–146 mm). At final time (24 h), sucrose is fully consumed by the enzyme. The degree of conversion of the acceptor (%) was calculated by
using the formula ([acceptor]initialꢀ[acceptor]final)/[sucrose]initial ꢃ100.
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 2288 – 2295 2291

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
tiopreference of NpAS. Indeed,
the results show that arabinose
and altrose are more efficiently
glucosylated by NpAS in their
l forms, whereas xylose is better
recognized in its d form. Inter-
estingly, glucosylation of man-
nose, xylose, and galactose ap-
peared to be less discriminant.
Conversely, the enzyme turned
out to be strictly enantiospecific
toward d-fucose. Altogether,
these results indicate that, in
spite of its broad acceptor pro-
miscuity, NpAS is able to discrim-
inate very similar molecules such
as enantiomers. In particular,
most recognized monosacchar-
ides, in either their d or l form,
such as d-fucose, d-allose, d-
xylose, d-mannose, l-altrose, and
Figure 5. Structure of the main products obtained by glucosylation of maltitol (13, 14), d-arabinose (15), and l-
arabinose (16, 17) by using NpAS as the catalyst and sucrose (1) as the donor of the glucosyl residue.
4
l-arabinose, are known to adopt predominantly a C1 ring con-
formation in solution, which is also the conformation adopted
by the natural acceptor substrate, d-glucose. The only excep-
tion found was d-galactose, which is the C4 epimer of d-glu-
cose. The reactivity of the OH-4 group in this substrate might
be impaired relative to the reactivity of this group in d-glu-
cose, and thus, spatial rearrangement of the monosaccharide
in the +1 subsite might be required. To investigate further the
origin of this selectivity, we thoroughly characterized the trans-
glucosylation products obtained from three distinct acceptors
(one of each group of molecules). The selected acceptors were
l-arabinose, d-arabinose, and maltitol. The reaction products
were purified from the reaction mixture by preparative chro-
matography. Analysis of the purified compounds by NMR spec-
troscopy was then performed to determine the structures of
the products. Interestingly, transglucosylation specificity was
found to be highly dependent on the nature of the acceptor
and original linkage specificities were observed (Figure 5).
+1 subsite may force the acceptor molecule to adopt the
most suitable conformation for productive catalysis. To further
investigate the formation of the glucosylated products ob-
tained with l-arabinose and d-arabinose, in silico analyses
were performed and compared to the crystallographic binding
modes of the natural products. With regard to the product
formed with d-arabinose, comparison of the docking modes
revealed that b-d-arabinofuranosyl-(1!1)-a-d-glucopyranoside
overlays better with sucrose than with maltose (Figure 6). d-
Arabinose in the furanosyl conformation was found to fit
better into the NpAS active site than in its pyranosyl conforma-
tion, especially given that the d-arabinopyranose prefers to
1
adopt a C4 conformation, in which the majority of the hydrox-
yl groups are oriented equatorially. Analysis of the docking
modes adopted by a-d-glucopyranosyl-(1!3)-l-arabinopyra-
nose revealed a 1208 rotation of the pyranose ring at the +1
1
Analysis of the compounds by H NMR spectroscopy confirmed
the presence of pure regioisomers in the glucosylation of both
d- and l-arabinose. The glucosylation of maltitol led to short-
chain oligosaccharides with sorbitol at the extremity, and the
glucosyl units are all linked through a-1,4 bonds. In the case of
the glucosylation of arabinose, the type of linkage was found
to be related to the ring conformation of the acceptor. Indeed,
structural characterization of the disaccharide products re-
vealed that when glucosylated, d-arabinose adopts a furanosyl
conformation, and the product was characterized as b-d-arabi-
nofuranosyl-(1!1)-a-d-glucopyranoside. Conversely, glucosyla-
tion of l-arabinose resulted in the formation of a-d-glucopyra-
nosyl-(1!3)-l-arabinopyranose with an equilibrium observed
between the a and b forms of the reducing end (65:35 in
D2O). The modification of linkage specificity may be attributed
to the conformation of the acceptor and the binding mode.
The active site topology is probably determinant for the ac-
commodation of the acceptor; in particular, the residues of the
Figure 6. Docking of the products obtained by glucosylation of d-arabinose
(blue) and l-arabinose (orange) overlaid with sucrose (green) and maltose
(purple), respectively. a) The stabilization of b-d-arabinofuranosyl-(1!1)-a-d-
glucopyranoside is preserved compared to that of sucrose and could explain
the preference for the furanosyl ring conformation of d-arabinose. b) The
polar network involved in a-d-glucopyranosyl-(1!3)-l-arabinopyranose
docking is increased relative to that of maltose docking and is probably re-
sponsible for the high glucosylation rate of l-arabinose.
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 2288 – 2295 2292

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
subsite. Comparison with crystallographic maltose [a-d-gluco-
pyranosyl-(1!4)-d-glucopyranose] revealed a nice superposi-
tion of the pyranosyl rings at the +1 subsite. A binding mode
with l-arabinopyranose conformation could be envisioned, in
which the 3-hydroxyl group occupies the same position as the
4-hydroxyl group of maltose, the natural product.
Experimental Section
Molecular modeling
3D models of the sucrose derivatives were constructed by using
the builder module of the InsightII software package (Accelrys, San
Diego, CA, USA).[38] Docking of the derivatives into the active site
of NpAS was performed by using the 3D structure of NpAS cocrys-
tallized with sucrose (PDB accession code 1JGI) as a template. All
complexes were subjected to energy minimization by using CFF91
force-field with a steepest descent algorithm (3000 iterations). En-
ergies of interaction (van der Waals, electrostatic, and total ener-
gies) were evaluated by using the docking module of InsightII.
Considering that in water solution, both d- and l-arabinose
predominantly adopt an a-pyranoid form,[34,35] it is interesting
to observe that NpAS prefers to act on d-arabinose in the b-
furanoid form, although present in very low amounts (2%) in
the tautomeric equilibrium. The X-ray structure of NpAS as
a complex with turanose indicated that recognition of the fruc-
tose tautomer at the +1 subsite was favored in linear form
rather than in the furanoid conformation because of the con-
strained topology of the active site.[36] The formation of b-d-
arabinofuranosyl-(1!1)-a-d-glucopyranoside by using d-arabi-
nose as the acceptor provides more evidence for the high con-
formational discrimination occurring at the +1 subsite. Alto-
gether, our results indicate that depending on the acceptor
used and its binding conformation in the NpAS active site, we
were able to broaden the regiospecificity of the enzyme. In ad-
dition to the glucosylated acceptors disclosed herein, we previ-
ously reported the successful glucosylation of N-acetyl-d-glu-
cosamine and l-rhamnose.[37] The various carbohydrate struc-
tures obtained by using NpAS revealed distinct but exclusive
formation of a-1-1-, a-1-3-, and a-1-4-linkages in useful rates
for synthetic purposes.
Expression and purification of NpAS
Expression and purification of recombinant glutathione (S)-transfer-
ase-amylosucrase fusion protein (GST-NpAS) were performed as
previously described.[39–41] The GST-NpAS protein was purified by
affinity chromatography by using glutathione-Sepharose 4B sup-
port (GE Healthcare). The GST tag was then removed by using
Prescission protease (GE Healthcare), which left five residues
(GPLGS) of the cutting site at the N-terminal extremity.
NpAS activity assay
One unit of amylosucrase activity corresponds to the amount of
enzyme that catalyzes the release of 1 mmol of reducing sugars per
minute in the assay conditions. NpAS specific activity was deter-
mined at 308C by using 50 gLꢀ1 sucrose in 50 mm Tris-HCl buffer,
pH 7.0. The concentration of the reducing sugars was determined
by using the dinitrosalicylic method and fructose as a standard.[42]
Synthesis of 3-ketosucrose
Conclusions
Sucrose was specifically dehydrogenated by using the pyranose
dehydrogenase from Agaricus meleagris (generous gift from Dr.
Clemens Peterbauer).[43] The reaction was performed in water at
258C with 1,4-benzoquinone as the proton acceptor. The total con-
version of sucrose into 3-ketosucrose (2) was observed after 24 h.
Its chemical structure was confirmed by NMR spectroscopy.[44]
13C NMR (100 MHz, D2O): d=207.25 (C3), 104.32 (C2’), 95.32 (C1),
81.83 (C5’), 76.47 (C3’), 75.87 (C5), 74.25 (C4’), 74.18 (C2), 71.67
(C4), 62.73 (C6’), 61.45 (C1’), 60.60 ppm (C6).
The donor promiscuity of NpAS was investigated by using 11
sucrose derivatives. The role of active site residues and the
high specificity of NpAS for sucrose have been underlined.
Among the tested molecules, p-Nitrophenyl-a-d-glucopyrano-
side was identified as the only alternative donor substrate for
NpAS and could be a relevant candidate for developing color-
based screening methods. The kinetic parameters of NpAS to-
wards this substrate were determined and were found to be
lower than those of sucrose. Thus, NpAS displays very narrow
donor substrate promiscuity, mainly because of its strong spe-
cificity for the glucosyl moiety at the ꢀ1 subsite. In addition,
a set of 20 hydroxylated molecules was evaluated as potential
acceptor substrates, and a broad specificity towards the ac-
ceptors was revealed. Of the 20 molecules assayed, 19 were
successfully recognized as acceptor substrates by NpAS. Re-
sults on l- and d- monosaccharides underlined the enantiopre-
ference of NpAS. Above all, this work emphasizes the dual evo-
lution of NpAS, which is both highly specific for the glucosyl
donor substrate sucrose and highly promiscuous towards
a wide range of hydroxylated acceptors. Both biochemical and
computational analyses gave new insights into NpAS struc-
ture–activity relationships, which are of prime interest in guid-
ing future experiments of NpAS evolution and in extending
the repertory of reactions catalyzed by this enzyme.
Synthesis of sucrose esters
Novozym 435 (lipase B from Candida antarctica) was used
(12.5 gLꢀ1) for the esterification of sucrose. The reaction was per-
formed in 2-methyl-2-butanol at 408C with vinyl acetate (4 equiv.)
as the acylating agent. Both sucrose-6-acetate (3) and sucrose-6,6’-
diacetate (4) were obtained and further chromatographed on silica
with dichloromethane/methanol (3.5:1) as the mobile phase. Their
structures were confirmed by NMR spectroscopy.[45]
Sucrose-6-acetate: 13C NMR (125 MHz, [D6]DMSO): d=170.47
(COCH3), 103.89 (C2’), 91.50 (C1), 82.80 (C5’), 77.04 (C3’), 74.59 (C4’),
72.77 (C3), 71.61 (C2), 70.27 (C5), 70.05 (C4), 63.90 (C6), 62.72 (C6’),
62.37 (C1’), 20.77 ppm (COCH3).
Sucrose-6,6’-diacetate: 13C NMR (125 MHz, C2D6SO): d=170.42
(COCH3), 170.28 (COCH3), 104.08 (C2’), 91.52 (C1), 79.06 (C5’), 76.40
(C3’), 74.55 (C4’), 72.63 (C3), 71.47 (C2), 70.40 (C4), 70.05 (C5), 65.81
(C6), 62.27 (C6’), 61.92 (C1’), 20.70 (COCH3), 20.61 ppm (COCH3).
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 2288 – 2295 2293

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
Synthesis of sucrose analogues
Synthesis, isolation, and characterization of the acceptor re-
action products
The sucrose analogues, namely, a-d-xylopyranosyl-(1–2)-b-d-fructo-
furanoside (XylFru, 6) and a-d-galactopyranosyl-(1–2)-b-d-fructofur-
anoside (GalFru, 5) were synthesized by using the commercial le-
vansucrase from Bacillus subtilis (from Johnson Matthey Cata-
lysts).[22,46] Sucrose (0.6m) was used as the fructosyl donor for the
glycosylation of xylose (1.2m) and galactose (1.2m) in phosphate
buffer (60 mm, pH 6.0) with conversions of 60 and 56%, respective-
ly. XylFru was purified from the reaction medium by preparative
chromatography (C-18 Luna). The reaction medium containing
GalFru and sucrose was submitted to the action of dextransucrase
DSRS vardel D4N.[47] The remaining sucrose was used as a glucosyl
donor for the synthesis of dextran and was removed by ethanol
precipitation. GalFru was thus purified by preparative chromatogra-
phy (K+). Pure XylFru and GalFru were characterized by NMR
spectroscopy.[46]
To characterize the different products resulting from the acceptor
reactions, three syntheses were carried out on a preparative scale.
The acceptor was chosen in each of three series: d-arabinose (d-
sugar series), l-arabinose (l-sugar series), and maltitol (polyol
series).
Preparative synthesis: Oligosaccharides issued from each acceptor
glucosylation were produced in a 100 mL batches containing su-
crose (146 mm) and acceptor (146 mm) by using purified NpAS
(1.5 Umlꢀ1) at 308C. All the acceptor reactions were stopped after
total sucrose consumption (24 h) by heating at 958C for 5 min.
Then, the reaction mixtures were centrifuged (4000 g, 20 min, 48C)
to remove proteins and filtered on a 0.22 mm membrane.
Preparative chromatography: The reaction products were separat-
ed by reverse-phase preparative chromatography by using a C18
column (Bischoff Chromatography) and detected with a refractome-
ter (Bischoff). Ultrapure water was used as the eluent at a constant
flow rate of 50 mLminꢀ1. Each peak corresponding to the main ac-
ceptor reaction products was separately collected, concentrated,
and reinjected into the analytical HPLC system described above to
ascertain purity. The structures of the products formed were identi-
fied by mass spectrometry and NMR spectroscopy.
GalFru: 13C NMR (100 MHz, D2O): d=106.2 (C1’), 94.91 (C1), 83.91
(C5’), 79.11 (C3’), 76.65 (C4’), 74.05 (C5), 71.74 (C3), 71.71 (C4), 70.59
(C2), 64.98 (C6’), 64.09 (C1’), 63.49 ppm (C6).
XylFru: 13C NMR (100 MHz, D2O): d=106.35 (C1’), 94.98 (C1), 84.01
(C5’), 78.78 (C3’), 76.23 (C4’), 75.38 (C3), 73.67 (C2), 71.81 (C4),
64.43 (C6’), 64.37 (C5), 63.43 ppm (C1’).
Main products obtained from a-d-glucopyranosyl-(1!4)-a-d-sorbi-
tol (maltitol) glucosylation:[48] a-d-Glucopyranosyl-(1!4)-a-d-gluco-
pyranosyl-(1!4)-a-d-sorbitol (S-G1-G2, 13): 13C NMR (100 MHz,
D2O): d=100.26 (C1G2), 99.66 (C2G1), 81.86 (C4S), 76.63 (C4G1), 73.28
(C2S), 72.84 (C3G2), 72.64 (C3G1), 72.57 (C5S), 71.71 (C5G1), 71.55
(C2G2), 71.46 (C3S), 70.91 (C2G1), 70.41 (C5G2), 69.28 (C4G2), 62.73
(C6S), 62.18 (C1S), 60.43 (C6G2), 60.33 ppm (C6G1). (a-d-Glucopyrano-
syl-(1!4)-a-d-glucopyranosyl-(1!4)-a-d-glucopyranosyl-(1!4)-a-
d-sorbitol (S-G1-G2-G3, 14): 13C NMR (100 MHz, D2O): d=100.25
(C1G3), 99.69 (C1G1), 99.56 (C1G2), 81.82 (C4S), 76.83 (C4G2),76.65
(C4G1), 73.32 (C2S), 72.97 (C3G1, C3G2), 72.83 (C3G3), 72.66 (C5S), 71.70
(C5G1, C5G2), 71.53 (C2G3), 71.50 (C3S), 71.27 (C2G1), 70.88 (C2G2),
70.42 (C5G3), 69.79 (C4G3), 62.73 (C6S), 62.18 (C1S), 60.43 (C6G3),
60.27 ppm (C6G1, C6G2).
Donor reactions with sucrose derivatives
Reactions were performed at 308C with the glycosyl donor
(150 mm, 30 mm for pNP-glycosides) in Tris-HCl buffer (50 mm),
pH 7.0 (6.0, 7.0, and 8.0 for G1P) with the enzyme (1 Umlꢀ1). After
24 h, the enzyme was inactivated at 958C for 10 min and removed
by centrifugation and filtration. The medium was then analyzed by
HPLC by using a Dionex P 680 series pump, a Shodex RI 101 series
refractometer, and an autosampler HTC PAL with K+ column
(Biorad, Aminex HPX-87K, 300ꢃ7.8 mm) or C-18-RPFusion (Phe-
nomenex, Synergi 4u Fusion-RP 80A, 250ꢃ4.60 mm, 4 m) for
pNP-glycosides.
Main product obtained from d-arabinose glucosylation: b-d-Arabi-
nofuranosyl-(1!1)-a-d-glucopyranoside (15):[49] 13C NMR (100 MHz,
D2O): d=101.97 (C1A), 99.70 (C1G), 81.50 (C4A), 76.43 (C2A), 73.89
(C3A), 72.79 (C3G), 72.55 (C5G), 71.48 (C2G), 69.29 (C4G), 62.99 (C5A),
60.18 ppm (C6G).
Acceptor reactions
As a first step, the reaction protocol was established in microplate
format to screen larger libraries of potential acceptors (hydroxylat-
ed compounds) for their capacity to be glucosylated by NpAS. This
study focused on 20 acceptors of interest: 14 monosaccharides (7
from the d series and 7 from the l series), and 6 polyols. Micro-
plate wells were filled with 200 mL reaction mixture containing su-
crose (146 mm), acceptor (292 mm), and Tris-buffer (50 mm, pH 7.5)
without enzyme (to represent the reaction at the initial time). In
parallel, a similar microplate was prepared with enzyme (to repre-
sent the reaction at the final time). Enzymes were used at 1 Umlꢀ1
to enable total sucrose consumption. Sucrose and acceptor were
introduced in a 1:2 ratio to favor acceptor glucosylation. After 24 h
incubation at 308C under agitation (100 rpm), the reaction mix-
tures were centrifuged (4000g, 5 min, 48C) and transferred to
a filter microplate (glass fiber membrane, PS, 0.25 mm pore) to be
clarified. Filtration was carried out by centrifugation of the filter mi-
croplate (5 min, 2000g, 48C) into a novel microplate for further
analysis. Efficiency of the glucosylation reaction was evaluated by
HPLC analysis of the acceptor reaction by using a Dionex P 680
series pump, a Shodex RI 101 series refractometer, and an auto-
sampler HTC PAL.
Main products obtained from l-arabinose glucosylation: a-d-Gluco-
pyranosyl-(1!3)-l-arabinopyranose in b form (16):[50–52] 13C NMR
(100 MHz, D2O): d=95.51 (C1G), 92.63 (C1A), 74.03 (C3A), 72.81
(C3G), 71.81 (C5G), 71.32 (C2G), 69.35 (C4G), 66.85 (C2A), 65.33 (C4A),
62.11 (C5A), 60.39 ppm (C6G). a-d-Glucopyranosyl-(1!3)-l-arabino-
pyranose in a form (17): 13C NMR (100 MHz, D2O): d=96.70 (C1A),
95.40 (C1G), 77.32 (C3A), 72.81 (C3G), 71.74 (C5G), 71.32 (C2G), 70.27
(C2A), 69.35 (C4G), 65.99 (C5A), 64.95 (C4A), 60.28 ppm (C6G).
Kinetics with p-nitrophenyl-a-d-glucoside
The hydrolytic activity of NpAS on p-nitrophenyl-a-d-glucoside was
determined by measuring the absorbance of p-nitrophenoxide at
405 nm and 308C by using a VersaMax Microplate Reader (Molecu-
lar Devices, Sunnyvale, CA, USA). The reactions were carried out in
Tris-buffer (50 mm, pH 7.0) by using purified NpAs (1 Umlꢀ1). A
range of concentrations from 1.5 to 30 mm of pNP-Glc was used
because of solubility. Concentration was related to absorbance at
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 2288 – 2295 2294

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
[24] M. H. Jensen, O. Mirza, C. Albenne, M. Remaud-Simeon, P. Monsan, M.
Gajhede, L. K. Skov, Biochemistry 2004, 43, 3104–3110.
[25] O. Mirza, L. K. Skov, M. Remaud-Simeon, G. P. de Montalk, C. Albenne, P.
Monsan, M. Gajhede, Biochemistry 2001, 40, 9032–9039.
405 nm by the Beer–Lambert law with extinction coefficientꢃ
sample cell length (el)=3.9 103 ꢃM-1 cm-1. The variation in the ab-
sorbance of the assay was monitored for 15 min.
[26] L. K. Skov, O. Mirza, A. Henriksen, G. P. De Montalk, M. Remaud-Simeon,
P. Sarcabal, R. M. Willemot, P. Monsan, M. Gajhede, Acta Crystallogr. Sect.
D 2000, 56, 203–205.
Acknowledgements
[27] L. K. Skov, O. Mirza, D. Sprogoe, B. A. Van der Veen, M. Remaud-Simeon,
C. Albenne, P. Monsan, M. Gajhede, Biocatal. Biotransform. 2006, 24,
99–105.
[28] F. Guꢀrin, S. Barbe, S. Pizzut-Serin, G. Potocki-Veronese, D. Guieysse, V.
Guillet, P. Monsan, L. Mourey, M. Remaud-Simeon, I. Andrꢀ, S. Tranier, J.
Biol. Chem. 2012, 287, 6642–6654.
[29] B. Henrissat, G. Davies, Curr. Opin. Struct. Biol. 1997, 7, 637–644.
[30] C. Albenne, G. P. De Montalk, P. Monsan, L. Skov, O. Mirza, M. Gajhede,
M. Remaud-Simeon, Biologia 2002, 57, 119–128.
D.D. was supported by a grant from the French Ministry of Re-
search. The authors would like to thank M.-S. Bounaix and A.-M.
BesanÅon for their technical assistance. The authors are also
grateful to Dr. C. Peterbauer for his generous gift of pyranose
dehydrogenase.
Keywords: enzymes
substrate promiscuity · transglucosylation
· glycosides · glycodiversification ·
[31] P. SarÅabal, M. Remaud-Simeon, R. M. Willemot, G. P. de Montalk, B.
Svensson, P. Monsan, FEBS Lett. 2000, 474, 33–37.
[32] T. P. Binder, J. F. Robyt, Carbohydr. Res. 1983, 124, 287–299.
[33] G. Potocki de Montalk, M. Remaud-Simeon, R. M. Willemot, P. Sarcabal,
V. Planchot, P. Monsan, FEBS Lett. 2000, 471, 219–223.
[1] U. T. Bornscheuer, R. J. Kazlauskas, Angew. Chem. 2004, 116, 6156–6165;
Angew. Chem. Int. Ed. 2004, 43, 6032–6040.
[34] S. J. Angyal, V. A. Pickles, Aust. J. Chem. 1972, 25, 1695–1710.
[35] O. M. Hanniffy, A. S. Shashkov, A. P. Moran, M. M. Prendergast, S. N. Sen-
chenkova, Y. A. Knirel, A. V. Savage, Carbohydr. Res. 1999, 319, 124–132.
[36] E. Champion, F. Guerin, C. Moulis, S. Barbe, T. H. Tran, S. Morel, K. De-
scroix, P. Monsan, L. Mourey, L. A. Mulard, S. Tranier, M. Remaud-
Simeon, I. Andrꢀ, J. Am. Chem. Soc. 2012, 134, 18677–18688.
[37] E. Champion, I. Andrꢀ, L. A. Mulard, P. Monsan, M. Remaud-Simeon, S.
Morel, J. Carbohydr. Chem. 2009, 28, 142–160.
[2] O. Khersonsky, D. S. Tawfik, Annu. Rev. Biochem. 2010, 79, 471–505.
[3] R. W. Gantt, P. Peltier-Pain, W. J. Cournoyer, J. S. Thorson, Nat. Chem.
Biol. 2011, 7, 685–691.
[4] G. J. Williams, C. Zhang, J. S. Thorson, Nat. Chem. Biol. 2007, 3, 657–
662.
[5] H. Hellmuth, L. Hillringhaus, S. Hobbel, S. Kralj, L. Dijkhuizen, J. Seibel,
ChemBioChem 2007, 8, 273–276.
[6] P. M. Coutinho, B. Henrissat, Recent Adv. Carbohydr. Bioeng. 1999, 3–12.
[7] I. Andrꢀ, G. Potocki-Veronese, S. Morel, P. Monsan, M. Remaud-Simeon,
Top. Curr. Chem. 2010, 294, 25–48.
[38] U. Dinur, A. Hagler, New approaches to empirical force fields, Wiley,
1991, p. 99–164.
[39] G. P. De Montalk, M. Remaud-Simeon, R. M. Willemot, V. Planchot, P.
Monsan, J. Bacteriol. 1999, 181, 375–381.
[8] E. Champion, I. Andrꢀ, C. Moulis, J. Boutet, K. Descroix, S. Morel, P.
Monsan, L. A. Mulard, M. Remaud-Simeon, J. Am. Chem. Soc. 2009, 131,
7379–7389.
[9] H. K. Cho, H. H. Kim, D. H. Seo, J. H. Jung, J. H. Park, N. I. Baek, M. J. Kim,
S. H. Yoo, J. Cha, Y. R. Kim, C. S. Park, Enzyme Microb. Technol. 2011, 49,
246–253.
[10] J. H. Jung, D. H. Seo, S. J. Ha, M. C. Song, J. Cha, S. H. Yoo, T. J. Kim, N. I.
Baek, M. Y. Baik, C. S. Park, Carbohydr. Res. 2009, 344, 1612–1619.
[11] D. H. Seo, J. H. Jung, S. J. Ha, H. K. Cho, D. H. Jung, T. J. Kim, N. I. Baek,
S. H. Yoo, C. S. Park, Appl. Microbiol. Biotechnol. 2012, 94, 1189–1197.
[12] D. H. Seo, J. H. Jung, S. J. Ha, M. C. Song, J. Cha, S. H. Yoo, T. J. Kim, N. I.
Baek, C. S. Park, J. Mol. Catal. B 2009, 60, 113–118.
[13] C. Albenne, L. K. Skov, O. Mirza, M. Gajhede, G. Potocki-Veronese, P.
Monsan, M. Remaud-Simeon, FEBS Lett. 2002, 527, 67–70.
[14] G. Okada, E. J. Hehre, Carbohydr. Res. 1973, 26, 240–243.
[15] G. Avigad, D. S. Feingold, S. Hestrin, J. Biol. Chem. 1957, 224, 295–307.
[16] J. Schneider, B. Hofer, FEBS J. 2007, 274, 373–373.
[17] B. Y. Tao, P. J. Reilly, J. F. Robyt, Carbohydr. Res. 1988, 181, 163–174.
[18] L. E. Baciu, H. J. Jordening, J. Seibel, K. Buchholz, J. Biotechnol. 2005,
116, 347–357.
[40] S. Emond, S. Mondeil, K. Jaziri, I. Andre, P. Monsan, M. Remaud-Simeon,
G. Potocki-Veronese, FEMS Microbiol. Lett. 2008, 285, 25–32.
[41] S. Pizzut-Serin, G. Potocki-Veronese, B. A. van der Veen, C. Albenne, P.
Monsan, M. Remaud-Simeon, FEBS Lett. 2005, 579, 1405–1410.
[42] J. B. Sumner, S. A. Howell, J. Biol. Chem. 1935, 108, 51–54.
[43] C. Sygmund, R. Kittl, J. Volc, P. Halada, E. Kubatova, D. Haltrich, C. K. Pe-
terbauer, J. Biotechnol. 2008, 133, 334–342.
[44] J. Volc, P. Sedmera, P. Halada, G. Daniel, V. Prikrylova, D. Haltrich, J. Mol.
Catal. B 2002, 17, 91–100.
[45] X. E. Yang, P. Zheng, Y. Ni, Z. H. Sun, J. Biotechnol. 2012, 161, 27–33.
[46] J. Seibel, R. Moraru, S. Gotze, Tetrahedron 2005, 61, 7081–7086.
[47] C. Moulis, A. Arcache, P. C. Escalier, M. Rinaudo, P. Monsan, M. Remaud-
Simeon, G. Potocki-Veronese, FEMS Microbiol. Lett. 2006, 261, 203–210.
[48] T. K. Kim, D. C. Park, Y. H. Lee, J. Microbiol. Biotechnol. 1997, 7, 310–317.
[49] L. A. M. van den Broek, E. L. van Boxtel, R. P. Kievit, R. Verhoef, G. Beld-
man, A. G. J. Voragen, Appl. Microbiol. Biotechnol. 2004, 65, 219–227.
[50] M. Doudoroff, W. Z. Hassid, H. A. Barker, J. Biol. Chem. 1947, 168, 733–
746.
[51] C. Goedl, T. Sawangwan, P. Wildberger, B. Nidetzky, Biocatal. Biotrans-
form. 2010, 28, 10–21.
[19] D. Daudꢀ, M. Remaud-Simeon, I. Andrꢀ, Nat. Prod. Rep. 2012, 29, 945–
960.
[52] W. Z. Hassid, M. Doudoroff et al., J. Am. Chem. Soc. 1948, 70, 306–310.
[20] A. Homann, J. Seibel, Appl. Microbiol. Biotechnol. 2009, 83, 209–216.
[21] J. Seibel, K. Buchholz, Adv. Carbohydr. Chem. Biochem. 2010, 63, 101–
138.
[22] J. Seibel, R. Moraru, S. Gotze, K. Buchholz, S. Na’amnieh, A. Pawlowski,
H. J. Hecht, Carbohydr. Res. 2006, 341, 2335–2349.
Received: January 9, 2013
[23] K. Demuth, H. J. Jçrdening, K. Buchholz, Carbohydr. Res. 2002, 337,
1811–1820.
Published online on May 24, 2013
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2013, 5, 2288 – 2295 2295

Copyright of ChemCatChem is the property of Wiley-Blackwell and its content may not be
copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.