Synthesis of derivatized chitooligomers using transglycosidases engineered from the fungal GH20 β-N-acetylhexosaminidase

Article abstract of DOI:10.1002/adsc.201500075

Abstract The synthesis of oligosaccharides using mutant glycosidases has been dynamically developing due to the need for novel carbohydrate-based materials. Chitooligomers (β-1→4-linked oligomers of N-acetylglucosamine) are bioactive compounds applicable in many industrial and pharmacological areas; however, their accessibility is still rather low. In this work, GH20 β-N-acetylhexosaminidase from the fungus Talaromyces flavus was engineered by site-directed mutagenesis to obtain three efficiently transglycosylating variants with ca. 200-times suppressed hydrolytic activity. Thus, we have prepared the first GH20 transglycosidases. In the reactions catalyzed by these mutant β-N-acetylhexosaminidases we were able to easily prepare and isolate both natural and modified chitooligomers in sufficient amounts for their complete spectral characterization and possible further application. The presented method for the synthesis of chitooligomers with aglycones suitable for linking to other biological structures is simple and robust enough to be easily scaled up.

Full text of DOI:10.1002/adsc.201500075

FULL PAPERS
DOI: 10.1002/adsc.201500075
Synthesis of Derivatized Chitooligomers using Transglycosidases
Engineered from the Fungal GH20 b-N-Acetylhexosaminidase
a,
a,b
a
a
´
Kristyna Slµmovµ, * Jana Krejzovµ, Petr Marhol, Lubica Kalachova,
c
d
e
a
ˇ
ˇ
Natallia Kulik, Helena Pelantovµ, Josef Cvacka, and Vladimír Kren
Laboratory of Biotransformation, Institute of Microbiology, Academy of Sciences of the Czech Republic, Vídenskµ 1083,
CZ 142 20 Prague, Czech Republic
a
ˇ
Fax: (+420)-296-442-509; phone: (+420)-296-442-510; e-mail: slamova.kristyna@gmail.com or slamova@biomed.cas.cz
Department of Biochemistry and Microbiology, Institute of Chemical Technology Prague, Technickµ 5, CZ 166 28 Prague,
Czech Republic
Department of Structure and Function of Proteins, Institute of Nanobiology and Structural Biology of GCRC, Academy
of Sciences of the Czech Republic, Zµmek 136, CZ 373 33 NovØ Hrady, Czech Republic
Laboratory of Molecular Structure Characterization, Institute of Microbiology, Academy of Sciences of the Czech
b
c
d
e
ˇ
Republic, Vídenskµ 1083, CZ 142 20 Prague, Czech Republic
Institute of Organic Chemistry and Biochemistry, Academy of Sciences of the Czech Republic, Flemingovo nµm. 2, CZ
166 10 Prague, Czech Republic
Received: January 26, 2015; Revised: March 27, 2015; Published online: May 12, 2015
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201500075.
Abstract: The synthesis of oligosaccharides using alyzed by these mutant b-N-acetylhexosaminidases
mutant glycosidases has been dynamically develop- we were able to easily prepare and isolate both natu-
ing due to the need for novel carbohydrate-based ral and modified chitooligomers in sufficient
materials. Chitooligomers (b-1!4-linked oligomers amounts for their complete spectral characterization
of N-acetylglucosamine) are bioactive compounds and possible further application. The presented
applicable in many industrial and pharmacological method for the synthesis of chitooligomers with agly-
areas; however, their accessibility is still rather low. cones suitable for linking to other biological struc-
In this work, GH20 b-N-acetylhexosaminidase from tures is simple and robust enough to be easily scaled
the fungus Talaromyces flavus was engineered by up.
site-directed mutagenesis to obtain three efficiently
transglycosylating variants with ca. 200-times sup- Keywords: b-N-acetylhexosaminidase; enzyme catal-
pressed hydrolytic activity. Thus, we have prepared ysis; glycosides; protein engineering; transglycosi-
the first GH20 transglycosidases. In the reactions cat- dases
Introduction
taining exo-glycosidases, in which the activated glyco-
side is employed. However, the concurrent hydrolysis
Carbohydrate chemistry has recently seen its boom, of both the substrates and the transglycosylation
mainly due to the need for novel carbohydrate-based products lowers the yields, which typically range be-
materials and the dynamic development of glycobiol- tween 20–40%. In the past decade, the classical enzy-
ogy and glycomics. Enzymatic synthesis of oligosac- matic synthesis of glycosides has shifted towards spe-
charides has often proved to be the method of choice, cialized applications, which brought along the need to
as it affords the desired products in a single step develop novel methodologies including substrate and
under mild conditions; moreover, the enzyme-cata- enzyme engineering (for more details please see
lyzed reactions are usually regio- and stereospecific. recent comprehensive reviews).[1–4]
The natural function of glycosidases is the hydrolysis
Glycosidases, well known for their substrate flexi-
of glycosidic bonds; however, under unnatural condi- bility, can be employed in the reactions with unnatu-
tions, glycosidic bonds can also be established by the ral glycosyl donors bearing various functional groups
action of glycosidases. Generally, the more common in their molecule.[5–8] Using site-directed mutagenesis,
reaction mode is transglycosylation, which is the ki- a new type of glycosidases, designated as glycosyn-
netically controlled process usually catalyzed by re- thases, was established in the late 1990s.[9,10] Glycosyn-
Adv. Synth. Catal. 2015, 357, 1941 – 1950
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1941

´
FULL PAPERS
Kristyna Slµmovµ et al.
thases, originally derived from the classical retaining when unnatural substrates were employed.[5,27,28] Few
exo-glycosidases, have the original active-site nucleo- b-N-acetylhexosaminidases are also able to catalyze
phile substituted by a non-nucleophilic amino acid, reversed hydrolysis yielding 1!6 linked chitooligom-
which causes knock-out of the hydrolytic activity ers.[29] Amongst the fungal enzymes, b-N-acetylhexos-
along with the strong preference for transglycosyla- aminidase isolated from Talaromyces flavus CCF2686
tion, thus creating a powerful biosynthetic tool. The (TfHex) found its prominent position due to its ex-
disadvantage of glycosynthases is that they require tremely broad substrate flexibility and great transgly-
glycosyl fluorides of the opposite anomeric configura- cosylation yields with a variety of modified sub-
tion than their original substrate as glycosyl donors, strates.[6,30] TfHex was recently sequenced (GenBank
which are not always easily accessible compounds. ID: JN601495) and a robust and high-yielding method
Recent progress in this field was achieved by the con- for its extracellular overexpression in the yeast Pichia
struction of a glycosynthase from the inverting glyco- pastoris was developed.[31] This progress enabled us to
sidase[11] and also from the endo-b-N-acetylglucosami- start designing functional mutants of this promising
nidase and chitinase, both employing the substrate-as- enzyme, aiming at decreasing its undesired hydrolytic
sisted catalysis.[12,13] On the other hand, transglycosi- activity while transglycosylation would be preferred,
dases, which are rare glycosidases strongly preferring thus the construction of a GH20 transglycosidase was
transglycosylation over hydrolysis while utilizing sub- the target of our efforts.
strates in their natural anomeric configuration, have
been described as both the wild-type[14] and as forms model of TfHex,[32] the active siteꢁs Tyr470 was select-
engineered from retaining glycosidases.[12,15,16]
ed as the promising amino acid residue to be mutated
Based on a recently published computational
Chitooligomers (b-1!4-linked oligomers of N-ace- to regulate hydrolytic activity of GH20 enzymes. Mu-
tylglucosamine) were found to possess beneficial bio- tation for Phe, His and Asn indeed brought the appar-
logical activities; e.g., they are potent antioxi- ent knock-out of the original hydrolytic activity and
dants,[17,18] antiangiogenic agents[19] and elicitors of the increased transglycosylation with substrates in the
immune response in plants.[20,21] Traditionally, chitoo- natural b-configuration. Using these engineered var-
ligomers have been produced by hydrolysis of chitin, iants of TfHex, we were able to prepare and isolate
which is a vastly abundant biowaste, either chemically a number of new derivatized chitooligomers suitable
using strong acids[18] or enzymatically by the action of for further application in biological studies. To the
chitinases.[22] Unfortunately, these robust methods best of our knowledge, we report here for the first
often lead to the generation of products which are dif- time a transglycosidase able to catalyze the produc-
ficult to isolate with and have a broad degree of poly- tion of rare higher chitooligomers and their glyco-
merization, while oligomers of a defined size are usu- sides.
ally desired. To solve this problem, enzymatic synthe-
sis using mutant chitinases has been explored recently,
however, as the reactions with the chitinase-derived Results and Discussion
glycosynthases required special donors (chitobiose-ox-
azoline[23] or chitobiose-a-fluoride[13,24]), all of the re- Preparation of Mutant Variants of b-N-
ported reactions were performed only on an analyti- Acetylhexosaminidase (TfHex)
cal scale and none of the proposed products was iso-
lated and spectrally identified.
In our previous work we have shown the versatile
b-N-Acetylhexosaminidases (EC 3.2.1.52; CAZy ability of b-N-acetylhexosaminidase from Talaromyces
GH20, http://www.cazy.org) are retaining exo-glycosi- flavus to efficiently catalyze the transglycosylation re-
dases catalyzing the cleavage of terminal b-d-GlcNAc actions,[6,27,30] unfortunately the yields of the transgly-
and b-d-GalNAc residues in various N-acetyl-b-d- cosylation products were always lowered by the con-
hexosaminides.[25] Like other glycosidases utilizing the current hydrolysis, which is a typical feature of glyco-
substrates bearing the 2-acetamido moiety, GH20 b- sylations performed by glycosidases. To overcome this
N-acetylhexosaminidases catalyze the reaction using problem of product hydrolysis, we decided to design
the substrate-assisted mechanism, in which this and prepare TfHex-based transglycosidases; function-
moiety is included in the catalysis instead of the en- al mutants of TfHex with diminished hydrolytic activi-
zymeꢁs catalytic nucleophile; the reaction intermedi- ty while the transglycosylation activity would be
ate in this scenario is the oxazoline form of the re- strongly preferred (Figure 1). Three most important
spective substrate instead of the classical covalent amino acids in the active site of the GH20 b-N-acetyl-
enzyme-substrate complex typical for other retaining hexosaminidases have been identified: the conserved
glycosidases.[26]
catalytic pair Asp-Glu and the tyrosine residue, which
b-N-Acetylhexosaminidases isolated from filamen- helps the catalytic aspartate to stabilize the oxazoline
tous fungi have been extensively studied due to their reaction intermediate.[33,34]
high natural transglycosylation abilities, especially
1942
asc.wiley-vch.de
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2015, 357, 1941 – 1950

FULL PAPERS
Synthesis of Derivatized Chitooligomers using Transglycosidases
Table 1. Specific activities of the wild-type and Tyr470 mu-
tants of b-N-acetylhexosaminidase from T. flavus.
Enzyme
variant
Specific activity
[U/mg]
Relative specific
activity [%]
wild type (wt)
MutYF
MutYH
95
100
0.5
0.7
3.5
0.52
0.74
3.68
MutYN
Figure 1. Transglycosylation reaction catalyzed by a transgly-
cosidase derived from the GH20 b-N-acetylhexosaminidase.
the closely related b-N-acetylglucosaminidases from
the family GH84.
The mutants were prepared by site-directed muta-
genesis of the TfHex gene integrated into the yeast
expression vector pPICZaA using the respective PCR
primers and a Pfu DNA polymerase, propagated in E.
coli. The positive clones were transformed and ex-
pressed in Pichia pastoris as described previously.[31]
As expected, the mutant b-N-acetylhexosaminidases
practically lost their hydrolytic activity, so their ex-
pression was monitored by SDS-PAGE of the concen-
trated culture media. In all three cases, the expression
and one-step purification on a cation-exchange resin
was achieved in the same quality and quantity as with
the wild-type enzyme, yielding ca. 10 mg of purified
protein from 100 mL of culture media after three
days upon induction with methanol (the SDS-PAGE
results of all purified TfHex variants are shown in
Figure S2 in the Supporting Information). The hydro-
lytic activity (measured as the hydrolysis of pNP-
GlcNAc) of the prepared mutants was strongly dimin-
ished compared to the wild-type enzyme ranging from
0.5 to 3.5%; the average specific activities of the indi-
vidual variants are shown in Table 1. Under the reac-
tion conditions for the measurement of hydrolytic ac-
tivity (2 mM substrate, 10 min), no transglycosylation
reaction occurs, which was checked experimentally
for all the enzyme variants. Moreover, all of the
mutant enzymes retained the long-term stability of
the parent enzyme; they can be stored as stock solu-
Figure 2. Model of the active site of TfHex with the NAG-
oxazoline reaction intermediate bound. Amino acid residues
important for catalysis: Asp370, Glu371 and Tyr470.
The model of the active site of TfHex with the tions at pH 5 and 48C for several months without any
amino acids important for the substrate binding and loss of activity, which is a great advantage for their
catalysis is shown in Figure 2 (in color in Figure S1 in potential application.
the Supporting Information); the computational
model of the b-N-acetylhexosaminidase from T. flavus
has been published recently.[32] For the design of the Transglycosylation Reactions using pNP-GlcNAc (1)
transglycosidases we were inspired by the work on as Donor and Acceptor Catalyzed by Mutant b-N-
the endo-b-N-acetylglucosaminidase, where the au- Acetylhexosaminidases
thors mutated the water-stabilizing tyrosine for phe-
nylalanine and created an endo-transglycosidase uti- The main goal of this work was to prepare GH20
lizing the N-glycan oxazoline sugar donors for the transglycosidases, so once we had the mutant b-N-ace-
biosynthesis of glycopeptides.[12] Analogously, we de- tylhexosaminidases in our hands, we focused on the
cided to change Tyr470 in TfHex for phenylalanine, transglycosylation abilities of these enzymes. At first,
and moreover, we opted for two more variants: histi- autocondensation reactions employing pNP-GlcNAc
dine, which is smaller and heteroaromatic, and aspar- (1) both as glycosyl donor and acceptor to optimize
agine, which is found in the homologous position in the reaction conditions on the analytical scale were
Adv. Synth. Catal. 2015, 357, 1941 – 1950
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1943

´
FULL PAPERS
Kristyna Slµmovµ et al.
Table 2. Isolated yields [mg] of transglycosylation reactions
catalyzed by the individual b-N-acetylhexosaminidase var-
iants (substrate pNP-GlcNAc 1, 80 mg per reaction, 6 h).
Enzyme
Product
1
2
3
4
5
6
wild type (wt) (30 mg) 26.6 3.8
2.6
0
16.3 3.2
MutYF (4.4 mg)
MutYH (1.0 mg)
18.5 9.4
29.1 11.8 4.4 4.8
9.3 14.0
0
0
0
0
6 h by boiling for 2 min to ensure the same conditions.
The reaction mixtures were purified using gel permea-
tion chromatography with water as the mobile phase,
all isolated compounds were lyophilized and their
structures were determined using NMR and MS (see
the Supporting Information).
The masses of the individual compounds obtained
from the reaction mixtures are given in Table 2; the
Tyr470Asn mutant was not used in this case, as it
strongly preferred the formation of the higher insolu-
ble products, which could not be separated and fully
characterized. The data in Table 2 clearly show the
Scheme 1. General scheme for a transglycosylation reaction
(substrate pNP-GlcNAc 1 acted both as a glycosyl donor
and acceptor) catalyzed by a b-N-acetylhexosaminidase.
performed; the progress of the reactions was moni- ability of the mutant b-N-acetylhexosaminidases to
tored by TLC. In these reactions, the formation of act as transglycosidases, as they readily catalyzed the
pNP-chitooligomers was observed within the first synthesis of pNP-chitooligomers in significant yields
hour, starting with pNP-chitobiose and building up and in a very simple reaction set-up while no hydroly-
the number of N-acetylglucosamine units in the prod- sis during the six-hour reaction time was detected and
ucts with increasing reaction time (Scheme 1, TLC no N-acetylglucosamine was isolated as a proof of
Figure S3 in the Supporting Information).
this statement. On the other hand, the yields of the
In the reactions catalyzed by any of the mutant b- transglycosylation reaction catalyzed by the wild-type
N-acetylhexosaminidase variants, we could detect the enzyme were apparently hampered by the coincident
production of the pNP-chitooligomers with up to six hydrolysis; a considerable amount of free N-acetylglu-
GlcNAc units after four hours, moreover, when the cosamine as well as a minor portion of chitobiose, the
Tyr470Asn mutant was used, even higher insoluble product of reverse hydrolysis of GlcNAc, were isolat-
oligomers were obtained, which we observed as the ed; moreover, the yields of the transglycosylation
formation of a white gel in the reaction mixture (TLC products were just minute. Altogether these results in-
is shown in Figure S4 in the Supporting Information). dicate that the prepared mutant b-N-acetylhexosami-
MALDI-TOF MS showed the presence of pNP-chi- nidases feature strongly improved synthetic abilities
tooligomers comprising at least seven N-acetylglucos- compared to the wild-type enzyme and proved to be
amine units; the full mass spectrum is shown in Fig- robust and suitable tools for application in the biosyn-
ure S5 in the Supporting Information. Importantly, in thesis of pNP-chitooligomers, which are rather rare
all of the reaction mixtures the presence of the hydro- and expensive chemicals.
lytic product N-acetylglucosamine was not observed;
this indicated that the mutation of Tyr470 for phenyl-
alanine, histidine and asparagine was a success in the Transglycosylation Reactions using pNP-GlcNAc (1)
preparation of a GH20 transglycosidase providing as Glycosyl Donor and GlcNAc-linker-t-Boc (7) as
hardly any detectable hydrolytic activity and also high Acceptor Catalyzed by Mutant b-N-
synthetic ability.
Acetylhexosaminidases
The optimized reaction conditions were scaled up
and the preparative transglycosylation reactions using After the successful synthesis of pNP-chitooligomers
the wild-type and mutant b-N-acetylhexosaminidases 2–4 in the simple reaction using pNP-GlcNAc, we de-
were performed to compare the composition of the cided to broaden the application of the mutant b-N-
products and their yields. All of the reactions were acetylhexosaminidases to the preparation of a new
started with 80 mg of substrate 1 (50 mM concentra- type of chitooligomers bearing the t-Boc moiety
tion) and the optimum amount of the individual linked to the reducing part of the glycoside via a thio-
enzyme was added; the reactions were stopped after urea-containing bridge. GlcNAc-linker-t-Boc (7,
1944
asc.wiley-vch.de
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2015, 357, 1941 – 1950

FULL PAPERS
Synthesis of Derivatized Chitooligomers using Transglycosidases
Scheme 2. Transglycosylation reaction catalyzed by mutant
b-N-acetylhexosaminidase using pNP-GlcNAc (1) as
a donor and GlcNAc-linker-t-Boc (7) as an acceptor.
Figure 3. HPLC chromatograms of the reaction mixtures
after 4 and 8 h; for the peak numbering see Scheme 1 and
Scheme 2.
Scheme 2) was previously synthesized in our laborato- ceptor 7 were equal (50 mM), after four hours when
ry and used as a base for the enzymatic synthesis of the donor was nearly depleted, another portion of the
LacNAc-type oligomers.[35] The advantage of this type donor (half of the original amount) was added and
of glycosides is the possibility to hydrolyze the t-Boc the reaction continued for another four hours
moiety and the released amino group at the end of (Figure 3). For this type of reaction the mutant
the flexible spacer can be used for simple linking of Tyr470Asn was identified as the most suitable cata-
respective chitooligomers to other structures, such as lyst, as it seemed to be most selective and effective in
multivalent glycodendrimers or surface materials.[36]
In the two-substrate transglycosylation reactions, 8–10.
the biosynthesis of the desired t-Boc-chitooligomers
their ratio is instrumental, as also the donor 1 can act
Under the above optimized reaction conditions, the
as an acceptor and thus two types of chitooligomers preparative transglycosylation reaction was performed
can be formed here (Scheme 2). The reaction condi- and its products were isolated using gel permeation
tions had to be finely tuned with respect to the con- chromatography and characterized (Table 3; NMR
centrations of the donor 1, acceptor 7 and also the and MS of the isolated products are given in the Sup-
enzyme variant used as the catalyst in order to shift porting Information). Thanks to the careful optimiza-
the composition of the mixture of products strongly tion of the conditions, the isolated products were only
towards the t-Boc-type of chitooligomers. The reac- the t-Boc-chitooligomers 8–10, only traces of the re-
tions were rather complex so that their progress could spective pNP-chitooligomers were detected by HPLC
not be monitored by TLC; therefore, a new HILIC- in the reaction mixture (Figure 3); these were com-
HPLC method for the detection of all substrates and pletely removed during the purification procedure. In
products contained in the reaction mixtures was de- this case we proved that the mutant b-N-acetylhexos-
veloped. Using this HPLC method, the optimum reac- aminidases with transglycosidase activity can be ap-
tion set-up leading to the maximum formation of the plied for the selective production of functionalized
required products was determined as follows: the chitooligoglycosides using the relatively cheap and
starting concentrations of both the donor 1 and ac- easily obtained substrate pNP-GlcNAc.
Adv. Synth. Catal. 2015, 357, 1941 – 1950
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1945

´
FULL PAPERS
Kristyna Slµmovµ et al.
Table 3. Isolated yields [mg] of the transglycosylation reac-
tion catalyzed by the Tyr470Asn mutant of TfHex (pNP-
GlcNAc 1, 75 mg total; GlcNAc-linker-t-Boc 7, 63 mg; 6 h).
mation of b-(1!6) linkages when the galacto- or 4-
deoxy substrates were used as acceptors.[28,30]
In our recent study we used the b-N-acetylhexos-
aminidase from T. flavus as a catalyst for the synthesis
of charged hexosaminides, the most suitable substrate
was found to be the pNP-GlcNAc-6-sulfate (12),
which was used both as a donor and acceptor, and the
mono- and di-sulfated dimeric products were obtained
in high yields.[6] Now, we were interested whether the
substrate 12 could be used for the synthesis of longer
sulfated chitooligomers, however, using the mutant b-
N-acetylhexosaminidases we obtained the same di-
Enzyme
Product
5
7
8
9
10
MutYN (1.0 mg)
0
28.3
21.0
14.2
8.1
Transglycosylation Reactions with Other b-N-
Acetylhexosaminidase Substrates
In our experiments with mutant b-N-acetylhexosam- meric product as with the wild-type enzyme. We
inidases we have also focused on a whole array of assume that the negative charge in the molecule of
substrates typically used with these enzymes that the product, which should be the acceptor in the sub-
were previously used as glycosyl donors (Figure 4). sequent reaction, did not allow proper binding in the
Even though the outcomes of these transglycosyla- active site of the enzyme and thus hindered the prog-
tions were not as successful as in the previous cases ress of the oligomerization reactions.
where pNP-GlcNAc (1) was employed as a donor, we
Glycosyl azides were introduced as efficient sub-
want to report on these reactions as well to show the strates for the transglycosylation reactions by our
complexity of the abilities of our mutant enzymes. group, namely the GlcNAc azide (13) was used for
These reactions were monitored by TLC; the data are the high-yielding synthesis of disaccharides catalyzed
not shown.
by the b-N-acetylhexosaminidase from T. flavus.[5]
At first we tested the C-4 epimer of the substrate 1, Surprisingly, when compound 13 was used as a sub-
e.g., pNP-GalNAc (11) as the donor in transglycosyla- strate with the mutant b-N-acetylhexosaminidases,
tion reactions; as the acceptor free N-acetylglucosa- neither hydrolysis nor transglycosylation occurred in
mine (5) was used in these reactions. We found that a significant manner; these mutants nearly completely
the galacto-configured substrate can be used as a gly- lost the ability to utilize the azido substrate 13 com-
cosyl donor, however, not as an acceptor, which led pared to the wild-type enzyme.
to the accumulation of the disaccharide product. Ap-
NAG-oxazoline (14) is the acid-labile transition
parently, the reaction could not proceed to the forma- state in the substrate-assisted catalytic mechanism
tion of higher oligomers as with the gluco-substrate 1. employed by the GH20 b-N-acetylhexosaminidases
This was rather surprising, as we had previously de- and other enzymes employing the substrates contain-
scribed the b-N-acetylhexosaminidase-catalyzed for- ing the 2-acetamido moiety.[26,33] The oxazoline-gly-
cans were used as glycosyl donors in the transglycosy-
lation reactions catalyzed by the wild-type and
mutant chitinase[23] and endo-b-N-acetylhexosamini-
dases.[37,38] This inspired us to test the GlcNAc-deriva-
tive NAG-oxazoline (14) as a donor in the reactions
catalyzed by the wild-type and mutant b-N-acetylhex-
osaminidases. As the oxazoline ring is very unstable
under mild acidic conditions, the reactions were per-
formed at pH 7.0, however, we could observe only the
spontaneous slow hydrolysis of the oxazoline moiety
leading to N-acetylglucosamine; no transglycosylation
products were detected with the wild-type and mutant
enzymes. These results were quite surprising as we as-
sumed that the reaction transition state would be
a preferable substrate, which was not the case.
Last but not least, we tested a preparativey trans-
glycosylation reaction, where the natural chitooligom-
er chitotriose (15; 50 mM, 150 mg; MutYH) – ob-
tained from the acid-catalyzed hydrolysate of chitin –
was employed as the substrate. Here we observed fast
Figure 4. Other substrates tested with mutant b-N-acetylhex-
disproportionation of the substrate, leading to the
generation of a complex mixture of chitooligomers
with the degree of polymerization from one (N-ace-
osaminidases: 11=pNP-GalNAc; 12=pNP-GlcNAc-6-sul-
fate; 13=GlcNAc-azide; 14=NAG-oxazoline; 15=chito-
triose.
1946
asc.wiley-vch.de
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2015, 357, 1941 – 1950

FULL PAPERS
Synthesis of Derivatized Chitooligomers using Transglycosidases
tylglucosamine) up to seven, as determined by TLC, Experimental Section
HPLC and MS (see Figures S24 and S25 in the Sup-
porting Information). The type of glycosidic linkages General Materials and Methods
in the oligosaccharides was finally determined by
All reagents, unless stated otherwise, were used as pur-
NMR measurements as described in the Experimental
Section. The magnitude of JH-1,H-2 =8.3 Hz, confirmed
also using J-resolved experiments, corresponds to the
chased without further purification. TLC was performed on
aluminium sheets precoated with Silica Gel 60 (F254 Merck,
DE); the spots were visualized by UV light (254 nm) and/or
by spraying with 5% H2SO4 in ethanol and charring. Flash
chromatography was performed with Merck silica 60 (230–
400 mesh) as a stationary phase.
A DNeasy Mini Plant Kit from Qiagen was used for
DNA isolation. A High Pure Plasmid Isolation Kit and Gen-
opur Plasmid Midi Kit (Roche, DE) were used for plasmid
isolation. The EasySelect Pichia Expression Kit and Zeocin
were purchased from Invitrogen (US). Oligonucleotide pri-
mers were obtained from Generi Biotech (CZ). The pGEM-
T Easy Vector System was purchased from Promega (US).
The Pfu Turbo DNA polymerase was purchased from Agi-
lent (US). The enzymes and buffers for DNA manipulations
were obtained from New England Biolabs (US). Media
components were from Oxoid (UK). All other chemicals, in-
cluding antibiotics and p-nitrophenyl 2-acetamido-2-deoxy-
b-d-glucopyranoside, were purchased from Sigma Aldrich
(CZ).
1
beta configuration. H and 13C NMR spectra are given
in Figures S26 and S27, respectively. Expansion of the
area with the anomeric carbons is given in the Fig-
ure S28 – all in the Supporting Information. Purifica-
tion of the individual oligomers by gel filtration was
not efficient enough, so we finally obtained four
mixed fractions eluted from the column (DP 7–5; DP
5–4; DP 4–3; DP 2–1). However, this method (when
using an efficient separation method) is simple and
robust and perfectly suitable for the synthesis of rare
and expensive higher chitooligomers at a larger scale
using the low-molecular-weight chitooligomers (or
even their mixtures) easily obtained from the bio-
waste, chitin.
The electrospray mass spectra were recorded using an
LTQ Orbitrap XL hybrid mass spectrometer (Thermo
Fisher Scientific, Waltham, US) equipped with an electro-
spray ion source. The mobile phase consisted of methanol/
water (4:1), flow rate 100 mLminÀ1, and the samples were in-
jected using a 2-mL loop. The mass spectra of positively
charged ions were internally calibrated using the sodium
adduct of diisooctyl phthalate ion (m/z=413.26623) as
a lock mass. The MALDI mass spectra were recorded using
an UltrafleXtreme (Bruker Daltonics, DE), equipped with
1 kHz smart beam II laser (355 nm). The samples (1 mL,
10 mgmLÀ1 in water) were mixed with 2,5-dihydroxybenzoic
acid (DHB) matrix solution (1 mL, 10 mgmLÀ1 in water), de-
posited on the MALDI plate and left to dry. External mass
calibration was done using the Peptide or Protein Calibra-
tion Standard supplied by Bruker Daltonics.
Conclusions
In this work we have succeeded to engineer the
fungal GH20 b-N-acetylhexosaminidase to obtain
three variants of the respective transglycosidase by
single amino acid mutations in the active site of the
original enzyme; where tyrosine 470 interacting with
the heterocyclic reaction intermediate was changed
for phenylalanine, histidine and asparagine. These
mutant b-N-acetylhexosaminidases were successfully
applied in the synthesis of two types of chitooligogly-
cosides, differing in the aglycone part of their mole-
cules, without any apparent hydrolysis of the products.
The products with the degree of polymerization of up
to 4 were purified and fully characterized; the
method of their synthesis is simple and robust enough
to be easily scaled up. A number of typical b-N-ace-
tylhexosaminidase substrates were tested as glycosyl
donors; however, the only compound suitable for the
preparation of higher chitooligomers appeared to be
pNP-GlcNAc (1). Moreover, when the natural hexo-
saminidase substrate – chitotriose – was employed as
a substrate, the formation of higher oligomers was
also observed. This is the first report on transglycosi-
dases derived from the exo-glycosidase employing the
substrate-assisted catalytic mechanism, moreover,
unlike in the previous works with mutant chitinases[13]
and endo-b-N-acetylglucosaminidases,[12] the products
of the transglycosylation reactions were isolated and
fully characterized, including NMR spectra.
NMR spectra were recorded on a Bruker Avance III
1
600 MHz spectrometer (600.23 MHz for H, 150.94 MHz for
13C) at 308C in D2O (99.98% D, ARMAR Chemicals, Dçt-
tingen, Switzerland). The residual signal of water was used
as an internal standard (dH =4.508 ppm). Carbon chemical
shifts in D2O were referenced to acetone (dC =30.50 ppm).
COSY, HSQC, HSQC-TOCSY, 1d-TOCSY and HMBC
NMR experiments were performed using the manufacturerꢁs
1
software. H NMR and 13C NMR spectra were zero filled to
fourfold data points and multiplied by window function
before Fourier transformation. The two-parameter double-
1
exponential Lorentz–Gauss function was applied for H to
improve resolution and line broadening (1 Hz) was applied
to get a better 13C signal-to-noise ratio. Chemical shifts are
given in d scale with digital resolution justifying the report-
ed values to three (dH) or two (dC) decimal places. In some
cases hydrogen chemical shifts were taken from the HSQC
and reported to two decimal places. The assignment of
proton spin systems was achieved by COSY, and then trans-
ferred to carbons by HSQC. The differentiation of individu-
al sugar units was in the case of overlapped proton spectra
Adv. Synth. Catal. 2015, 357, 1941 – 1950
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1947

´
FULL PAPERS
Kristyna Slµmovµ et al.
supported also by 1d-TOCSY and HSQC-TOCSY experi-
ments. The anomeric configuration was determined from the
magnitude of JH-1,H-2. The glycosidic linkage position was de-
duced using heteronuclear correlations in HMBC and from
the downfield glycosylation shifts of the involved carbons C-
4. Several signals of protons and carbons adjacent to the thi-
ourea moiety were broadened due to the isomerism on the
as described previously.[31] Protein concentration was deter-
mined using the Bradford assay.[39]
The hydrolytic activity of the wild-type and mutant b-N-
acetylhexosaminidases was assayed spectrophotometrically
using p-nitrophenyl 2-acetamido-2-deoxy-b-d-glucopyrano-
side as a chromogenic substrate, starting concentration
2 mM. One unit of enzymatic activity was defined as the
amount of enzyme releasing 1 mmol of p-nitrophenol per
minute in 50 mM citrate-phosphate buffer at pH 5.0 and
358C. After incubation of the reaction mixture at 358C for
10 min, the liberated p-nitrophenol was determined spectro-
photometrically at 420 nm under alkaline conditions (50 mL
of the reaction mixture was added to 1 mL of 0.1M
Na2CO3). Under the reaction conditions for the measure-
ment of hydrolytic activity, no transglycosylation reaction
occurs, which was checked experimentally (TLC) for all the
enzyme variants.
À
NH C=S bond.
Site-Directed Mutagenesis of b-N-
Acetylhexosaminidase from Talaromyces flavus
The b-N-acetylhexosaminidase from Talaromyces flavus
CCF 2686 (TfHex) was sequenced (GenBank ID:
JN601495), cloned and expressed in Pichia pastoris in our
previous work.[31] The expression construct pPICZaA-
TfHex containing the TfHex gene was used as template for
the site-directed mutagenesis. Three mutants were prepared
using the following pairs of PCR primers:
HPLC – Hydrophilic Interaction Liquid
Chromatography (HILIC)
Tyr470Phe:
and 5’-GTAGCCGCAGTCGAGGAACATAAAGTCTGA-3’;
Tyr470His: 5’-TCAGACTTTATGCACCTCGACTGCGGCTAC-3’
and 5’-GTAGCCGCAGTCGAGGTGCATAAAGTCTGA-3’;
Tyr470Asn: 5’-TCAGACTTTATGAACCTCGACTGCGGCTAC-3’
5’-TCAGACTTTATGTTCCTCGACTGCGGCTAC-3’
Hydrophilic interaction liquid chromatography (HILIC) was
used to separate polar and hydrophilic compounds. HPLC
analyses were performed on a Shimadzu Prominence LC an-
alytical system consisting of a Shimadzu LC-20AD binary
HPLC pump, a Shimadzu SIL-20ACHT cooling autosam-
pler, a Shimadzu CTO-10AS column oven and a Shimadzu
SPD-20MA diode array detector (Shimadzu, JP). The
sample was dissolved in acetonitrile/water (3/1; v/v) and sep-
arated on TSKgel Amide-80 HILIC column (250”4.6 mm,
Tosoh Bioscience, DE). Binary gradient elution was used:
mobile phase A=100% acetonitrile; mobile phase B=
water; gradient, 0–7 min, 22% B; 7–20 min, 22–35% B; 20–
25 min, 35% B; 25–28 min, 35–22% B; and 28–33 min, 22%
B. Flow rate=1 mLminÀ1 at 278C; detection at 200 nm; in-
jection volume=1 mL.
Semi-preparative HPLC analyses were performed on
a Shimadzu Prominence LC system consisting of a Shimadzu
LC-20AD binary HPLC pump, a Shimadzu SIL-20ACHT
cooling autosampler, a Shimadzu CTO-10AS column oven,
a Shimadzu fraction collector FRC-10A and Shimadzu SPD-
20MA diode array detector (Shimadzu, JP). The sample was
dissolved in acetonitrile/water (3/1; v/v) and separated on
TSKgel Amide-80 HILIC column (300”7.8 mm, Tosoh Bio-
science, DE). Binary gradient elution was used: mobile
phase A=100% acetonitrile; mobile phase B=water; gradi-
ent, 0–7 min, 22% B; 7–20 min, 22–35% B; 20–25 min, 35%
B; 25–28 min, 35–22% B; and 28–33 min, 22% B. Flow
rate=3 mLminÀ1 at 278C; detection at 200 nm; injection
volume=50 mL.
and 5’-GTAGCCGCAGTCGAGGTTCATAAAGTCTGA-3’.
The site-directed mutagenesis PCR reactions were per-
formed in a T-Personal Thermal Cycler (Biometra, DE)
using Pfu Turbo DNA polymerase (Agilent, US). The PCR
conditions were as follows: initial denaturation at 958C for
1 min, followed by 30 cycles of denaturation at 958C for
30 s, primer annealing at 578C for 1 min, extension at 728C
for 6 min, with a final extension of 10 min at 728C. After
PCR 1 mL of DpnI (New England Biolabs, US) was added
directly to the reaction mixture (50 mL) to cleave the tem-
plate plasmid DNA, and the reaction was incubated over-
night at 378C. The PCR product was then transformed into
Escherichia coli JM109 competent cells using zeocin
(100 mgmLÀ1) as a selection marker; the plasmids isolated
from the resultant E. coli colonies were sequenced using the
internal sequencing primer 5’-TCAACTCGACCCGCTCAACAA
CAAGACGT-3’.
Expression and Purification of Mutant b-N-
Acetylhexosaminidases
The plasmids containing the correctly mutated TfHex gene
were propagated in E. coli JM109, isolated and used for
transformation of Pichia pastoris KM71H, where the
mutant enzymes were overexpressed using the same proto-
col as with the wild-type enzyme; all details of cultivation
procedures and media are given in Ref.[31] As the mutant en-
zymes lost most of their hydrolytic activity, their extracellu-
lar production by P. pastoris was monitored by 10% SDS-
PAGE of 10-fold concentrated expression media (2 and 3
days after the first induction by methanol). The best produc-
ing clones were cryopreserved at À808C and used for the
large scale production of the mutant enzymes, the yields of
protein were analogous to those obtained with the wild-type
enzyme (ca. 30–40 mg protein/200 mL culture after 3 days
after induction). The produced mutant enzymes were puri-
fied in a single step using cation-exchange chromatography
Analytical Transglycosylation Reactions
For autocondensation reactions using pNP-GlcNAc 1 both
as glycosyl donor and acceptor, the substrate (50–100 mM)
was suspended in 50 mM citrate-phosphate buffer pH 5.0.
The concentration of the wild-type and mutant b-N-acetyl-
hexosaminidases and the reaction time were optimized; the
reaction mixtures were incubated at 358C (1000 rpm) and
the progress of the reaction was monitored by TLC (pro-
pane-2-ol:H2O:NH3 aq., 7:2:1, v/v/v).
For the glycosylation of GlcNAc-linker-t-Boc (7)[35] using
pNP-GlcNAc (1) as the donor, both substrates (pNP-
1948
asc.wiley-vch.de
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2015, 357, 1941 – 1950

FULL PAPERS
Synthesis of Derivatized Chitooligomers using Transglycosidases
GlcNAc 50–120 mM; GlcNAc-linker-t-Boc 30–50 mM) were
suspended in 50 mM citrate-phosphate buffer pH 5.0 and
various amounts of mutant b-N-acetylhexosaminidases were
added. The reaction mixtures were incubated at 358C
(1000 rpm), at certain time intervals aliquots were taken
(50 mL of reaction mixture were added to 150 mL of acetoni-
trile and centrifuged for 10 min) and analyzed by HPLC as
described above.
to seven sugar units (the spectrum is shown in the Support-
ing Information, Figure S4).
Transglycosylation Reaction with b-N-Acetylhexos-
aminidase Mutant Y470N and pNP-GlcNAc (1) as
a Glycosyl Donor and GlcNAC-linker-t-Boc (7) as an
Acceptor
The donor pNP-GlcNAc (1; 50 mg, 0.147 mmol, 50 mM) and
acceptor GlcNAc-linker-t-Boc (7; 63 mg, 0.149 mmol,
50 mM) were dissolved in citrate-phosphate buffer pH 5.0
(2.5 mL), 0.8 mg (2.8 U) of the b-N-acetylhexosaminidase
mutant Y470N were added and the reaction volume was ad-
justed at 3.0 mL with the buffer. The reaction was incubated
at 358C with shaking (1000 rpm) and the progress of the re-
action was monitored by HILIC HPLC as described in the
separate section above. After 4 h of incubation when the
donor 1 was nearly consumed, another portion of 1 (25 mg,
0.073 mmol) was added and the reaction was continued for
4 more hours. After 8 h of total reaction time the reaction
was stopped by boiling for 3 min, centrifuged after cooling
to room temperature (13000 rpm for 10 min) and the super-
natant was loaded onto a Biogel P-2 column (2”100 cm,
mobile phase water, 10 mLhÀ1; Bio-Rad, US). Fractions
containing products were collected and lyophilized, all of
the products were obtained as white solids. The compounds
isolated from the reaction mixture were as follows:
GlcNAc-linker-t-Boc (7; 28.3 mg), chitobiose-linker-t-Boc
(8; 21.0 mg), chitotriose-linker-t-Boc (9; 14.2 mg) and chito-
tetraose-linker-t-Boc (10; 8.1 mg). As an alternative, the
same reaction was repeated and the products were isolated
using semipreparative HILIC HPLC as described above. We
obtained the same products, however, with lower purity. The
mass and NMR spectra (and data in tables) of the products
obtained by gel filtration are given in the Supporting Infor-
mation.
Synthesis of Chitooligosaccharides by the Wild-Type
and Mutant b-N-Acetylhexosaminidases
General Procedure using pNP-GlcNAc (1) as a Substrate:
Substrate 1 (80 mg, 0.235 mmol, 50 mM) was dissolved in
50 mM citrate-phosphate buffer pH 5.0 (3.0 mL), the opti-
mum amount of the wild-type or mutant enzyme was added
and the final reaction volume was adjusted to 4.69 mL with
the buffer. The reaction was incubated at 358C with shaking
(1000 rpm) and the progress of the reaction was monitored
by TLC (propan-2-ol:H2O:NH3 aq., 7:2:1, v/v/v). The reac-
tion was stopped by boiling for 3 min, centrifuged after cool-
ing to room temperature (13000 rpm for 10 min) and the su-
pernatant was loaded onto a Biogel P-2 column (2”100 cm,
mobile phase water, 10 mLhÀ1; Bio-Rad, US). Fractions
containing products were collected and lyophilized, all of
the products were obtained as white solids. The MS and
NMR spectra in extenso and data in tables of the products
are given in the Supporting Information.
Transglycosylation Reaction with pNP-GlcNAc (1) and
Wild-Type b-N-Acetylhexosaminidase: For this reaction
25 mg (2 U) of the wt b-N-acetylhexosaminidase were used
(general conditions as above) and the reaction was stopped
after 6 h. The compounds isolated from the reaction mixture
were: pNP-GlcNAc (1; 26.6 mg), pNP-chitobiose (2;
3.8 mg), pNP-chitotriose (3; 2.6 mg), N-acetylglucosamine
(GlcNAc, 5; 16.3 mg) and chitobiose (6; 3.2 mg).
Transglycosylation Reaction with pNP-GlcNAc (1) and
Mutant b-N-Acetylhexosaminidase Y470F: For this reac-
tion 4.4 mg (3.3 U) of the b-N-acetylhexosaminidase mutant
Y470F were used and the reaction (general conditions as
above) was stopped after 6 h. The compounds isolated from
the reaction mixture were as follows: pNP-GlcNAc (1;
18.5 mg), pNP-chitobiose (2; 9.4 mg), pNP-chitotriose (3;
9.3 mg) and pNP-chitotetraose (4; 14.0 mg).
Transglycosylation Reaction with pNP-GlcNAc (1) and
Mutant b-N-Acetylhexosaminidase Y470H: For this reac-
tion 1 mg (0.6 U) of the b-N-acetylhexosaminidase mutant
Y470H was used and the reaction (general conditions as
above) was stopped after 6 h. The compounds isolated from
the reaction mixture were: pNP-GlcNAc (1; 29.1 mg), pNP-
chitobiose (2; 11.8 mg), pNP-chitotriose (3; 4.4 mg) and
pNP-chitotetraose (4; 4.8 mg).
Acknowledgements
This work was supported by the Czech Science Foundation
grant 13-06818P (K.S.), bilateral CZ-DE Academic Project
M200201204 (V. K.) and by the Institutional research concept
(Institute of Microbiology, Prague) RVO61388971. The au-
ˇ
thors thank Dr. Alena Krenkovµ for the HPLC analysis of
chitooligomer mixtures.
References
Transglycosylation Reaction with pNP-GlcNAc (1) and
Mutant b-N-Acetylhexosaminidase Y470N: For this semi-
preparative reaction 4 mg of pNP-GlcNAc (0.012 mmol,
50 mM, total reaction volume 235 mL) and 50 mg (0.2 U) of
the b-N-acetylhexosaminidase mutant Y470N were used and
the reaction was stopped after 6 h. After centrifugation the
supernatant containing the soluble products was discarded
and the insoluble products were air-dried overnight; this
mixture of higher oligomers was analyzed by MALDI-TOF
MS and was shown to contain pNP-chitooligomers with up
ˇ
[1] P. Bojarovµ, V. Kren, Trends Biotechnol. 2009, 27, 199–
209.
[2] L.-X. Wang, W. Huang, Curr. Opin. Chem. Biol. 2009,
13, 592–600.
[3] T. Desmet, W. Soetaert, P. Bojarovµ, V. Kren, L. Dij-
khuizen, V. Eastwick-Field, A. Schiller, Chem. Eur. J.
2012, 18, 10786–10801.
[4] Z. Armstrong, S. G. Withers, Biopolymers 2013, 99,
666–674.
ˇ
Adv. Synth. Catal. 2015, 357, 1941 – 1950
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1949

´
FULL PAPERS
Kristyna Slµmovµ et al.
[5] P. Fialovµ, A. T. Carmona, I. Robina, R. Ettrich, P. Sed-
[22] C.-J. Woo, H.-D. Park, Biotechnol. Lett. 2003, 25, 409–
412.
ˇ
ˇ
ˇ
mera, V. Prikrylovµ, L. Petrµskovµ-Husµkovµ, V. Kren,
Tetrahedron Lett. 2005, 46, 8715–8718.
[23] J. Sakamoto, T. Watanabe, Y. Ariga, S. Kobayashi,
Chem. Lett. 2001, 1180–1181.
[24] T. Ohnuma, T. Fukuda, S. Dozen, Y. Honda, M. Kitao-
ˇ
ˇ
[6] P. Bojarovµ, K. Slµmovµ, K. Krenek, R. Gazµk, N.
Kulik, R. Ettrich, H. Pelantovµ, M. Kuzma, S. Riva, D.
ˇ
ˇ
Adµmek, K. Bezouska, V. Kren, Adv. Synth. Catal.
2011, 353, 2409–2420; Corrigendum: Adv. Synth. Catal.
2014, 356, 259.
ka, T. Fukamizo, Biochem. J. 2012, 444, 437–443.
ˇ
[25] K. Slµmovµ, P. Bojarovµ, L. Petrµskovµ, V. Kren, Bio-
technol. Adv. 2010, 28, 682–693.
[26] B. L. Mark, M. N. G. James, Can. J. Chem. 2002, 80,
1064–1084.
[7] G. Lopez, C. Nugier-Chauvin, C. Remond, M. O’Dono-
hue, Carbohydr. Res. 2007, 342, 2202–2211.
[8] A. Schroven, S. Meinke, P. Ziegelmuller, J. Thiem,
Chemistry 2007, 13, 9012–9021.
ˇ
[27] P. Fialovµ, D. J. Namdjou, R. Ettrich, V. Prikrylovµ, J.
ˇ
Rauvolfovµ, K. Krenek, M. Kuzma, L. Elling, K. Be-
[9] L. F. Mackenzie, Q. P. Wang, R. A. J. Warren, S. G.
Withers, J. Am. Chem. Soc. 1998, 120, 5583–5584.
[10] C. Malet, A. Planas, FEBS Lett. 1998, 440, 208–212.
[11] Y. Honda, S. Fushinobu, M. Hidaka, T. Wakagi, H.
Shoun, H. Taniguchi, M. Kitaoka, Glycobiology 2008,
18, 325–330.
[12] M. Umekawa, W. Huang, B. Li, K. Fujita, H. Ashida,
L. X. Wang, K. Yamamoto, J. Biol. Chem. 2008, 283,
4469–4479.
ˇ
ˇ
zouska, V. Kren, Adv. Synth. Catal. 2005, 347, 997–
1006.
ˇ
ˇ
[28] L. Weignerovµ, P. Vavruskovµ, A. Pisvejcovµ, J. Thiem,
V. Kren, Carbohydr. Res. 2003, 338, 1003–1008.
[29] E. Rajnochovµ, J. Dvorµkovµ, Z. Hunkovµ, V. Kren,
Biotechnol. Lett. 1997, 19, 869–872.
[30] K. Slµmovµ, R. Gazµk, P. Bojarovµ, N. Kulik, R. Et-
trich, H. Pelantovµ, P. Sedmera, V. Kren, Glycobiology
2010, 20, 1002–1009.
[31] K. Slµmovµ, P. Bojarovµ, D. Gerstorferovµ, B. Flie-
drovµ, J. Hofmeisterovµ, M. Fiala, P. Pompach, V.
Kren, Protein Expr. Purif. 2012, 82, 212–217.
[32] N. Kulik, K. Slµmovµ, R. Ettrich, V. Kren, BMC Bioin-
ˇ
ˇ
ˇ
ˇ
ˇ
ˇ
[13] E. A. Martinez, H. Boer, A. Koivula, E. Samain, H.
Driguez, S. Armand, S. Cottaz, J. Mol. Catal. B: Enzy-
matic 2012, 74, 89–96.
ˇ
ˇ
ˇ
[14] D. Simcíkovµ, M. Kotik, L. Weignerovµ, P. Halada, H.
ˇ
ˇ
Pelantovµ, K. Adamcovµ, V. Kren, Adv. Synth. Catal.
formatics 2015, 16, 28.
2015, 357, 107–117.
[33] I. Tews, A. C. Terwisscha van Scheltinga, A. Perrakis,
K. S. Wilson, B. W. Dijkstra, J. Am. Chem. Soc. 1997,
119, 7954–7959.
[34] B. L. Mark, D. J. Vocadlo, S. Knapp, B. L. Triggs-Raine,
S. G. Withers, M. N. G. James, J. Biol. Chem. 2001, 276,
10330–10337.
[15] F. Arab-Jaziri, B. Bissaro, M. Dion, O. Saurel, D. Harri-
son, F. Ferreira, A. Milon, C. Tellier, R. FaurØ, M. J.
O’Donohue, New Biotechnol. 2013, 30, 536–544.
[16] D. Teze, J. Hendrickx, M. Czjzek, D. Ropartz, Y.-H. Sa-
nejouand, V. Tran, C. Tellier, M. Dion, Protein Eng.
Des. Sel. 2014, 27, 13–19.
ˇ
[35] B. Sauerzapfe, K. Krenek, J. Schmiedel, W. W. Wa-
[17] D.-N. Ngo, S.-H. Lee, M.-M. Kim, S.-K. Kim, J. Funct.
ˇ
karchuk, H. Pelantovµ, V. Kren, L. Elling, Glycoconju-
gate J. 2009, 26, 141–159.
[36] C. Rech, R. R. Rosencrantz, K. Krenek, H. Pelantovµ,
Foods 2009, 1, 188–198.
[18] D.-N. Ngo, M.-M. Kim, S.-K. Kim, Carbohydr. Polym.
2008, 74, 228–234.
ˇ
ˇ
P. Bojarovµ, C. E. Rçmer, F.-G. Hanisch, V. Kren, L.
[19] Z. Wang, L. Zheng, S. Yang, R. Niu, E. Chu, X. Lin,
Biochem. Biophys. Res. Commun. 2007, 357, 26–31.
[20] M. Okada, M. Matsumura, Y. Ito, N. Shibuya, Plant
Cell Physiol. 2002, 43, 505–512.
[21] H. Kaku, Y. Nishizawa, N. Ishii-Minami, C. Akimoto-
Tomiyama, N. Dohmae, K. Takio, E. Minami, N. Shi-
buya, Proc. Natl. Acad. Sci. USA 2006, 103, 11086–
11091.
Elling, Adv. Synth. Catal. 2011, 353, 2492–2500.
[37] M. Fujita, S. Shoda, K. Haneda, T. Inazu, K. Takegawa,
K. Yamamoto, Biochim. Biophys. Acta 2001, 1528, 9–
14.
[38] H. Oichiai, W. Huang, L.-X. Wang, Carbohydr. Res.
2009, 344, 592–598.
[39] M. Bradford, Anal. Biochem. 1976, 72, 248–254.
1950
asc.wiley-vch.de
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2015, 357, 1941 – 1950