Charged hexosaminides as new substrates for β-N-acetylhexosaminidase- catalyzed synthesis of immunomodulatory disaccharides

Article abstract of DOI:10.1002/adsc.201100371

This work is a structure-activity relationship study that investigates the influence of the nature and amount of negative charge in carbohydrate substrates on the affinity of β-N-acetylhexosaminidases, and on the stimulation of natural killer cells. It describes synthetic procedures yielding novel glycosides that are useful in immunoactivation. Specifically, we present a thorough study on the ability of six C-6 modified β-N-acetylhexosaminides (aldehyde, uronate, 6-O-sulfate, 6-O-phosphate) to serve as substrates for cleavage and glycosylation by a library of β-N-acetylhexosaminidases from various sources. Four novel disaccharides with one or two (negatively) charged groups were prepared in synthetic reactions in good yields. Surprisingly, the 6-O-phosphorylated substrate, although cleaved by a number of enzymes from the series, worked neither as a donor nor as an acceptor in transglycosylation reactions. The results of wet experiments were supported by molecular modeling of substrates in the active site of two representative enzymes from the screening. All ten prepared compounds were examined in terms of their immunoactivity, namely as ligands of two activation receptors of natural killer (NK) cells, NKR-P1 and CD69, both with isolated proteins and whole cells. Sulfated disaccharides in particular acted as very efficient protectants of NK cells against activation-induced apoptosis, and as stimulants of the natural killing of resistant tumor cells, which makes them good candidates for potential clinical use in cancer treatment. Copyright

Full text of DOI:10.1002/adsc.201100371

FULL PAPERS
DOI: 10.1002/adsc.201100371
Charged Hexosaminides as New Substrates for
b-N-Acetylhexosaminidase-Catalyzed Synthesis of
Immunomodulatory Disaccharides
a,f
a,b,f
a
a
ˇ
ˇ
Karel Krenek, Radek Gazꢀk,
´
Pavla Bojarovꢀ, Kristyna Slꢀmovꢀ,
Natallia Kulik,c Rꢁdiger Ettrich,c Helena Pelantovꢀ,a Marek Kuzma,a Sergio Riva,d
a,e
a,e
a,
ˇ
ˇ
David Adꢀmek, Karel Bezouska, and Vladimꢂr Kren *
a
ˇ
Institute of Microbiology, Academy of Sciences of the Czech Republic, Vꢂdenskꢀ 1083, CZ-142 20 Praha 4, Czech
Republic
Fax : (+420)-296-442-509; phone: (+420)-296-442-510; e-mail: kren@biomed.cas.cz
Department of Biochemistry and Microbiology, Institute of Chemical Technology, Prague, Technickꢀ 5, CZ-16628 Praha
6, Czech Republic
Department of Structure and Function of Proteins, Institute of Nanobiology and Structural Biology of GCRC, Zꢀmek
136, CZ-373 33 Novꢃ Hrady, Czech Republic
Istituto di Chimica del Riconoscimento Molecolare, CNR, Via Mario Bianco 9, I-20131 Milano, Italy
Department of Biochemistry, Faculty of Science, Charles University in Prague, Hlavova 8, CZ-128 40 Praha 2, Czech
Republic
b
c
d
e
f
P. Bojarovꢀ and K. Slꢀmovꢀ contributed equally to this work.
Received: May 11, 2011; Revised: August 9, 2011; Published online: August 30, 2011
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/adcs.201100371.
Abstract: This work is a structure-activity relation- reactions. The results of wet experiments were sup-
ship study that investigates the influence of the ported by molecular modeling of substrates in the
nature and amount of negative charge in carbohy- active site of two representative enzymes from the
drate substrates on the affinity of b-N-acetylhexos- screening. All ten prepared compounds were exam-
ACHTUNGTRENNUNGaminidases, and on the stimulation of natural killer ined in terms of their immunoactivity, namely as li-
cells. It describes synthetic procedures yielding novel gands of two activation receptors of natural killer
glycosides that are useful in immunoactivation. Spe- (NK) cells, NKR-P1 and CD69, both with isolated
cifically, we present a thorough study on the ability proteins and whole cells. Sulfated disaccharides in
of six C-6 modified b-N-acetylhexosaminides (alde- particular acted as very efficient protectants of NK
hyde, uronate, 6-O-sulfate, 6-O-phosphate) to serve cells against activation-induced apoptosis, and as
as substrates for cleavage and glycosylation by a li- stimulants of the natural killing of resistant tumor
brary of b-N-acetylhexosaminidases from various cells, which makes them good candidates for poten-
sources. Four novel disaccharides with one or two tial clinical use in cancer treatment.
(negatively) charged groups were prepared in syn-
thetic reactions in good yields. Surprisingly, the 6-O-
phosphorylated substrate, although cleaved by a Keywords: b-N-acetylhexosaminidase; biotransfor-
number of enzymes from the series, worked neither mations; charged glycosides; glycosylation; molecu-
as a donor nor as an acceptor in transglycosylation lar modeling; natural killer cells
Introduction
synthetic potential;[2] at the same time, they are also
extensively studied from the viewpoint of their struc-
The employment of enzymes in carbohydrate synthe- ture and their involvement in human physiology and
sis has seen dynamic development in the past few de- disease.[3] Previously, we have demonstrated that
cades. b-N-Acetylhexosaminidases[1] (EC 3.2.1.52, fungal b-N-acetylhexosaminidases possess a wide tol-
CAZy GH20; http://www.cazy.org) belong to the erance to substrates bearing various modifications, in-
group of glycosidases with a well-documented broad cluding N-acyls,[4] sulfates,[5] and an absent C-4 hy-
Adv. Synth. Catal. 2011, 353, 2409 – 2420
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2409

Pavla Bojarovꢀ et al.
FULL PAPERS
droxy.[6] Moreover, GlcNAc substrates with two un-
ꢀ
usual C N linked aglycones, namely the azido
group[7] and the aromatic triazole moiety,[8] were
cleaved by some fungal b-N-acetylhexosaminidases;
the azide even worked as an efficient glycosyl donor
in transglycosylation reactions.[7]
As for the biological and especially the immunomo-
dulatory activity of glycosides and glycoconjugates,
our previous studies described the ability of N-acetyl-
d-hexosamine structures to stimulate the immune re-
sponse: they can activate natural killer (NK) cells by
binding to their activation receptors, such as NKR-P1
and CD69 proteins.[9] A significantly enhanced bind-
ing affinity was observed in structures bearing nega-
tively charged functional groups such as carboxyl or
Figure 1. Substrates 2–6 and 8.
sulfate,[9a,10] but no systematic study on this phenom- Figure 1). The results are supported by molecular
enon has been presented until now. The pilot study modeling using two enzymes from the set. Four novel,
on C-6 oxidized p-nitrophenyl 2-acetamido-2-deoxy- negatively charged disaccharides were prepared in
b-d-galacto-pyranosides[11,12] revealed the ability of b- high yields. The immunomodulatory properties of all
N-acetylhexosaminidases, especially those of the Peni- ten C-6-modified hexosaminides were tested using
cillium and Talaromyces genera, to cleave and trans- NK cell activation receptors NKR-P1A and CD69
fer the glycosyls of sugar aldehydes. After in situ oxi- and CD69high lymphocytes. As a result, we could com-
dation to a carboxyl group, the uronate-containing pare the influence of the number and character of the
oligosaccharides b-d-GalpNAcA-(1!4)-d-GlcpNAc functionalities in the hexosaminides on the synthetic
and b-d-GalpNAcA-(1!4)-b-d-GlcpNAc-(1!4)-d- performance of b-N-acetylhexosaminidase and on the
ManpNAc were prepared. These compounds are response in immunological tests.
among the best known oligosaccharidic ligands of the
CD69 receptor described so far. Another biological
property studied in this type of compounds is their Results and Discussion
ability to protect NK cells against activation-induced
apoptosis.[13]
Preparation of Substrates 2–6 and 8
In this work we report on the substrate specificity
and synthetic potential of (mainly fungal) b-N-acetyl- The synthesis of aldehyde 2 (Scheme 1) was adapted
hexosaminidases with C-6 modified p-nitrophenyl 2- from an elegant published procedure based on Dess–
acetamido-2-deoxy-b-d-hexopyranosides (aldehyde 2, Martin oxidation.[14] NMR spectroscopy confirmed
uronate 3, sulfates 4 and 6, phosphates 5 and 8; the formation of the required aldehyde together with
Scheme 1. Synthesis of b-N-acetylhexosaminidase substrates 2 (A), 3 (B) and 5 (C).
2410
asc.wiley-vch.de
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 2409 – 2420

Charged Hexosaminides as New Substrates for b-N-Acetylhexosaminidase-Catalyzed Synthesis
Table 1. Hydrolysis of substrates 2–5 by b-N-acetylhexosami-
its hydrated form (geminal diol). Upon standing in
anhydrous solution (DMSO), a gradual dehydration
of the diol was observed, which is in agreement with
the literature.[15,16] Uronate 3 was prepared in a high-
yield, one-step reaction using the laccase from Tra-
metes pubescens (EC 1.10.3.2) in tandem with
TEMPO. The enzymatic oxidation proceeded via al-
dehyde 2; however, it could not be efficiently stopped
at that stage. The aldehyde traces (<5%) could be
isolated by silica gel chromatography (EtOAc:
MeOH:H2O 6:8:0.25!6:8:1) but since the aldehyde
yield was negligible anyway, a more convenient sepa-
ration by gel permeation chromatography in water
was opted for, which gave pure uronate 3. Sulfates 4
(gluco-) and 6 (galacto-) were prepared in a 5-step
synthesis starting from the respective unprotected p-
nitrophenyl hexosaminide as already described in our
previous work.[6] The preparation of phosphates 5
(gluco-) and 8 (galacto-) was the most challenging
task and was complicated by the presence of the nitro
moiety on the aromatic ring. Here, the optimal proce-
dure was the phosphorylation of the respective par-
tially protected p-nitrophenyl hexosaminide with
POCl3. Product purification was particularly challeng-
ing as this compound tended not to separate from the
phosphate salts that originated in the reaction mixture
due to its polarity.
nidases.[a,b]
Enzyme source
2
3
4
5
Acremonium persicinum CCF
1850
ꢀ
ꢀ
ꢀ
ꢀ
A. awamori CCF 763[c]
A. flavofurcatis CCF 3061
A. flavus CCF 642
+ +
+ + +
+
+ +
+
+ + +
+ +
+ +
+ +
+ +
+ +
ꢀ
+
ꢀ
ꢀ
ꢀ
+
+
ꢀ
ꢀ
ꢀ
ꢀ
+
+
+
+
+
+
ꢀ
-
ꢀ
-
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
+ +
+
+
+
+
+
+
ꢀ
A. niger CCIM K2
ꢀ
A. niveus CCF 3057
A. nomius CCF 3086
A. oryzae CCF 147
A. oryzae CCF 1066
A. parasiticus CCF 1298
A. tamarii CCF 3085
A. versicolor CCF 2491
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
Fusarium oxysporum CCF 377 + +
+
P. brasilianum CCF 2171[c]
P. chrysogenum CCF 1269
P. oxalicum CCF 1959
P. oxalicum CCF 2315
P. oxalicum CCF 2430
P. pittii CCF 2277
+ +
+ +
+ +
+ +
+
+
+ +
+
+ +
+
ꢀ
+ + + + +
+ +
+ + + + + + +
+ +
+ +
+ +
T. flavus CCF 2573[c]
T. flavus CCF 2686
+ + + + + + +
+ +
ꢀ
T. ohiensis CCF 2229
T. striatus CCF 2232
Trichoderma harzianum CCF
2687
+
+
+
+
+ + +
+
ꢀ
+
Canavalia ensiformis[d]
Streptococcus pneumoniae
Bovine kidney
+ + +
+
+ +
+ +
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
+
ꢀ
+
b-N-Acetylhexosaminidase-Catalyzed Hydrolysis of
Compounds 2–6, 8 and Molecular Modeling
+ +
+
Human placenta (HEX A)
[a]
Substrates 6 and 8 were not cleaved by any of the en-
zymes (hydrolysis rate <1%).
(ꢀ) <2%, (+) 3–10%, (+ +) 11–20%, (+ + +) 21–30%
hydrolysis rate relative to the hydrolysis of standard sub-
strate pNP-GlcNAc (1).
All the prepared substrates, i.e., aldehyde 2, uronate
3, sulfates 4 and 6 and phosphates 5 and 8, were
tested with a series of 24 fungal b-N-acetylhexosami-
nidases produced by mycelia stemming from the Cul-
ture Collection of Fungi (CCF), Charles University in
Prague, CZ. In addition, four representatives from
other sources were tested. As shown in Table 1, the
best results were obtained with the fungal enzymes.
Aldehyde 2 proved to be the best substrate, its hy-
[b]
[c]
[d]
A.=Aspergillus, P.=Penicillium, T.=Talaromyces.
Canavalia ensiformis=jack bean.
The screening results were compared to the conclu-
drolysis rates being 10–30% of those measured with sions of computational docking[17] and molecular dy-
the standard substrate 1. In the gluco-series, the abili- namics simulations (MD) using two enzymes from the
ty of enzymes to cleave the substrates declined in the series: the b-N-acetylhexosaminidase from Penicillium
order aldehyde!sulfate!uronate!phosphate.
oxalicum,[18] which has a broad substrate specificity[4]
In contrast, the galacto-derivatives 6 (sulfate) and 8 and cleaved all tested gluco-substrates well, and the
(phosphate) were not cleaved by any enzymes from enzyme from Aspergillus oryzae,[19] a more conserva-
the set. This result corresponds to the previous con- tive hexosaminidase with a stricter substrate specifici-
clusions observed with p-nitrophenyl 2-acetamido- ty.
2-deoxy-b-d-galacto-pyranosiduronic
acid
(not
In both enzymes, DGs of the equilibrated enzyme-
cleaved).[11] Apparently, the introduction of a charged substrate complex with aldehyde 2 were comparable
group at the C-6 position of p-nitrophenyl hexosami- to the standard substrate 1 (for details see Table S1 in
nides is only tolerated by b-N-acetylhexosaminidases the Supporting Information). Uronate 3 provided
if the substrate is in the gluco-configuration (equatori- weaker interaction with both enzymes due to the loss
al C-4 hydroxy group). This behavior is consistent ir- of hydrogen bonds fixing C-5 and C-6 carbons in the
respective of the nature of the charged group in the substrate pyranosyl ring. The strongly charged sulfate
C-6 position (uronate, sulfate or phosphate).
4 and phosphate 5 in the P. oxalicum enzyme showed
Adv. Synth. Catal. 2011, 353, 2409 – 2420
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2411

Pavla Bojarovꢀ et al.
FULL PAPERS
just slightly lower DG values than the standard sub- biose-related disaccharide carrying one charged group
strate 1 and they could well accommodate in the at the non-reducing end (Scheme 3). All the reactions
active site. Therefore, it could be expected that these were first tested on an analytical scale in miligram
substrates were cleaved by the enzyme, though possi- amounts to optimize the amount of enzyme, reactant
bly with lower affinity. In contrast, phosphate 5 in the concentrations, reaction time, etc., and then repeated
A. oryzae enzyme initiated strong conformational at a hundred mg scale to gain the disaccharidic prod-
changes within the active site. During MD the sub- ucts in a sufficient amount for biological studies. The
strate C-4 hydroxy caused a strong shift and deforma- catalyst used in all the preparative reactions present-
tion of the loop Pro301–Gln312 above the active site ed was the b-N-acetylhexosaminidase from Talaromy-
(for reasoning and additional illustrating material see ces flavus CCF 2686, which proved to be among the
the Supporting Information), leading unambigously to most efficient enzymes in the screening and, addition-
the conclusion that the A. oryzae enzyme could not ally, was known for its good synthetic capacity and
accommodate phosphate 5 in its native active site.
tolerance to substrate modifications, as shown previ-
ously.[6,7,11]
The carboxyl group was introduced into the disac-
charidic compounds in two ways – via aldehyde 2 in
combination with in situ quantitative oxidation of the
Syntheses Catalyzed by b-N-Acetylhexosaminidases
Substrates 2–5, exhibiting good to excellent results in product by NaClO2 or directly via uronate 3. This was
the screening for cleavage, were used in two types of because, from our previous experience with galacto-
synthetic transglycosylation reactions. In the first derivatives,[11] we (correctly) supposed uronate 3 to be
design, they served both as glycosyl donor and accept- a much poorer substrate than aldehyde 2, both in
or, resulting in a pNP-chitobiose type b
G
disaccharide [b-d-GlcpNAc-(1!4)-b-d-GlcpNAc-(1! trast to the galacto-series where the uronate was not
O)-pNP] carrying two charged groups (Scheme 2). In cleaved at all and the only way to the carboxy disac-
the other design, GlcNAc (11) as acceptor was glyco- charides was through the aldehyde, uronate 3 was a
sylated with a modified glycosyl, leading to a chito- relatively good substrate for cleavage. However, the
Scheme 2. Autocondensation reactions with substrates 3 and 4 catalyzed by the b-N-acetylhexosaminidase from T. flavus
CCF 2686.
Scheme 3. Glycosylation of GlcNAc (11) using substrates 2 and 4 catalyzed by the b-N-acetylhexosaminidase from T. flavus
CCF 2686.
2412
asc.wiley-vch.de
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 2409 – 2420

Charged Hexosaminides as New Substrates for b-N-Acetylhexosaminidase-Catalyzed Synthesis
Table 2. Affinity of carbohydrate ligands to NK cell activa-
synthesis and subsequent complicated purification
(preparative HPLC) resulted in quite a poor synthetic
yield (16%). In contrast, aldehyde 2 gave ca. double
the yield in the synthetic reaction. The in situ oxida-
tion step of the product to a uronic acid with NaClO2
was very straightforward, selective and efficient. Thus,
we could conclude that it is easier to produce carboxy
disaccharides using an aldehyde donor.
Sulfate 4 was the best glycosyl donor in the series,
affording products 12 and 13 in very good yields by a
simple purification with gel permeation chromatogra-
phy (28 and 33%, respectively, after two hours of re-
action time). The procedure described here gave sub-
stantially better results than those by Uzawa and co-
workers,[20] especially as far as reaction time and
amount of enzyme are concerned. Unfortunately, we
were unable to prepare any transglycosylation prod-
ucts from phosphate 5. Neither of the reaction designs
described above gave any notable amounts of phos-
phorylated disaccharide products, although the cata-
lytic function of the enzyme was apparently un-
changed since b-d-GlcpNAc-(1!6)-b-d-GlcpNAc
could be isolated from the reaction with acceptor 11.
No result was obtained, even when phosphate 5 was
used as reaction acceptor with pNP-GlcNAc as donor.
Instead of the expected phosphorylated product, we
only isolated pNP-chitobiose. These results clearly
tion receptors NKR-P1A and CD69, expressed on a loga-
rithmic scale (ꢀlog IC50 ꢁSD). The values reflect the com-
pound ability to inhibit binding between the respective pro-
tein and a standard high affinity ligand GlcNAc23BSA.
Compound
NKR-P1A
CD69
GlcNAc (11)[a]
pNP-GlcNAc (1)
Aldehyde 2
Uronate 3
Sulfate 4
5.2ꢁ0.2
6.0ꢁ0.1
8.7ꢁ0.1
7.7ꢁ0.1
9.8ꢁ0.1
8.5ꢁ0.1
6.4ꢁ0.1
6.1
3.4ꢁ0.5
4.1ꢁ0.1
6.6ꢁ0.0
6.7ꢁ0.1
8.6ꢁ0.2
6.6ꢁ0.1
4.3ꢁ0.1
5.0
Phosphate 5
pNP-GalNAc
pNP-GalNAc aldehyde[b]
pNP-GalNAcA[b]
Sulfate 6
6.8
6.9
9.7ꢁ0.1
8.7ꢁ0.1
9.7ꢁ0.1
9.6ꢁ0.4
9.2ꢁ0.4
8.5ꢁ0.1
8.1ꢁ0.1
8.8ꢁ0.1
8.6ꢁ0.0
6.8ꢁ0.1
7.5ꢁ0.1
9.6ꢁ0.4
9.3ꢁ0.3
6.3ꢁ0.2
6.3ꢁ0.1
7.8ꢁ0.1
Phosphate 8
pNP-chitobiose
Disaccharide 9 (diuronate)
Disaccharide 12 (disulfate)
Chitobiose
Disaccharide 10 (uronate)
Disaccharide 13 (sulfate)
[a]
GlcNAc (positive control) had IC50 values of 6.3ꢅ10ꢀ6
(NKR-P1A) and 4.0ꢅ10ꢀ4 M (CD69).
Data taken from ref.[11]
M
[b]
show that phosphate 5, although cleaved, does not ligand strength (although any charged group causes
work as a substrate in transglycosylation reactions.
an increase in binding).
Prepared Compounds as Ligands of NK Cell
Activation Receptors
Immunoactivity of Prepared Compounds
In order to investigate the immunoactivity of the pre-
All prepared compounds were tested in a competitive pared compounds 2–6, 8, 9, 10, 12 and 13, we first
binding assay with two activation receptors of NK tested their effect on the activation of immune cells
cells, rat NKR-P1 and human CD69 (see Experimen- carrying cellular forms of the NKR-P1 and CD69 re-
tal Section for details).[21] d-Mannose was used as the ceptors, namely with highly NKR-P1 positive rat NK
negative control, no inhibitory effects could be ob- cells and CD69high human peripheral blood mononu-
served up to 10 mM concentrations (not shown). The clear cells (PBMC).[21] In both cases, the dicarboxylat-
results are summarized in Table 2 as ꢀlog IC50. In the ed disaccharide 9 and both sulfated disaccharides 12
monosaccharidic series, the introduction of an alde- and 13 displayed the highest effect. This correlated
hyde or a charged group into the molecule produced well with the competitive studies shown in Table 2.
a notable increase in binding, more pronounced with We also tested the ability of individual compounds to
the NKR-P1 receptor. Sulfates 4 and 6 were the best protect CD69high cells from apoptosis.[13] Again, com-
ligands in the series, followed by phosphates, uronates pounds 9, 12 and 13 were the best in this assay, fol-
and aldehydes. Generally, disaccharides carrying a lowed by monosulfated pNP-GlcNAc and pNP-
pNP moiety were better ligands than the reducing dis- GalNAc 4 and 6. Finally, the immunomodulatory
accharides. In the disaccharidic series, outstanding re- properties of the compounds were tested using human
sults were observed with the CD69 receptor. Two PBMC and human tumor cells, both NK-sensitive
charged groups in compounds 9 and 12 (carboxy and (K562), and NK-resistant (RAJI). Here, compounds
sulfate, respectively) produced an increase of two 9, 12 and 13 strongly stimulated the natural killing of
orders of magnitude compared to unmodified pNP- RAJI tumor cells. Such an outstanding ability to en-
chitobiose. With ꢀlog IC50 values approaching 10 they hance natural killing of normally resistant tumor cells
ranked among the best ligands ever studied, compara- may result from a synergic activation of NK cells
ble only to multivalent structures.[22] Apparently, the through NKR-P1 and their protection against apopto-
sulfate is the most efficient moiety for increasing sis, as shown in separate experiments above.
Adv. Synth. Catal. 2011, 353, 2409 – 2420
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2413

Pavla Bojarovꢀ et al.
FULL PAPERS
Conclusions
a Bruker AVANCE III 400 MHz spectrometer (400.13 MHz
for 1H, and 100.61 MHz for 13C), and compound 10 in a
Bruker AVANCE III 600 MHz spectrometer (600.23 MHz
This study presents a series of C-6-modified substrates
derived from pNP-GlcNAc and pNP-GalNAc and
studies their ability to act as substrates for a range of
b-N-acetylhexosaminidases in cleavage and novel syn-
thetic reactions. The compounds from the gluco-series
were much better cleaved than those from the galac-
to-series; aldehyde and sulfate proved to be the best
substrates. Interestingly, phosphate 5, although
cleaved by b-N-acetylhexosaminidases, did not work
as a glycosyl donor or acceptor in transglycosylation
reactions. Using the other substrates, four disacchar-
ides in total were enzymatically synthesized. Further-
more, we compared the prepared compounds in terms
of their immunomodulatory potential, namely their
ability to serve as ligands of NKR-P1A and CD69 ac-
tivation receptors and as protectants of CD69high lym-
phocytes against activation-induced apoptosis. Sul-
fates 4 and 6 were the best activating monosacchar-
ides, especially for the NKR-P1 receptor. The positive
effect of two charged groups in the (disaccharidic)
molecule was clearly visible in their reaction with the
CD69 receptor (compounds 9, 12). In the immunostu-
dies with whole cells, compounds 9, 12 and 13 strongly
stimulated the natural killing of NK-resistant RAJI
tumor cells.
In conclusion, sulfates were the most efficient com-
pounds in the series, both as substrates (in terms of
enzymatic cleavage and synthesis) and as the most
active immunostimulants of natural killer cells. These
compounds are especially valuable due to the favora-
ble combination of high affinity and good stability,
which distinguishes them from the previously tested
natural sulfated oligosaccharides.[10,23] These features
make these compounds promising lead structures for
the further development of carbohydrate mimetics
with potent immunomodulatory activities and the en-
zymatic synthetic methodology makes them available
for further studies.
for H, and 150.93 MHz for 13C) in DMSO-d6 (99.9 atom%
1
D, Sigma–Aldrich, Steinheim, D) or D2O (100 atom% D,
Sigma–Aldrich, Steinheim, D) at 303 K. The residual signal
of the solvents was used as an internal standard (DMSO-d6:
dH =2.500, dC =39.60, D2O dH =4.508). The carbon spectra
in D2O were referenced to the signal of acetone (dc =30.50).
1H NMR, 13C NMR, COSY, HSQC, HMQC, HMBC, and
1D-TOCSY spectra were measured using standard manufac-
turersꢆ software. The proton spin systems of each sugar unit
were assigned by COSY and 1D-TOCSY. The assignment
was transferred to carbons by HSQC. The position of sugar
moiety substitution was confirmed by long-range heteronu-
clear correlations in the HMBC spectrum and the configura-
tion at C-1 was deduced from coupling constants JH1,H2 ob-
1
served in the H NMR spectra. The position of the sulfate
was indicated indirectly by the downfield shifts of the in-
volved carbons (C-6). Chemical shifts are given on a d scale
[ppm], and coupling constants in Hz. Digital resolution ena-
bled us to report the chemical shifts of protons to three,
coupling constants to one and carbon chemical shifts to two
decimal places. Some hydrogen or chemical shifts were read
from HSQC or HMQC and are reported to two decimal
places.
Mass spectrometry analysis and MS/MS experiments were
performed using a LCQDECA ion trap mass spectrometer
(ThermoQuest, San Jose, CA) equipped with a static nanoe-
lectrospray ion source. The spray voltage was maintained at
1.42 kV, the tube lens voltage was ꢀ10 V. The heated capil-
lary was kept at 1808C with a voltage of ꢀ37 V. Positive-ion
full scans were acquired over the m/z range 150–2000. To
confirm the structure of analyzed compound the MS/MS ex-
periments were performed with normalized collision energy
in range 20–35%, activation Q 0.25, activation time 30 ms
and with isolation width m/z 3.0. MALDI-TOF mass spectra
were measured on an ultraFLEX matrix-assisted laser de-
sorption/ionization reflectron time-of-flight mass spectrome-
ter (Bruker-Daltonics, Bremen, D). The ion acceleration
voltage was 19 kV and the reflectron voltage was set to
20 kV. Positive spectra were calibrated externally using the
monoisotopic [M+H]+ ion of human angiotensin I m/z=
1296.69, MRFA peptide m/z=524.26 and CCA matrix peak
m/z=379.09. A saturated solution of a-cyano-4-hydroxycin-
namic acid or 2,5-dihydrobenzoic acid in 50% MeCN/0.3%
acetic acid was used as the MALDI matrix. 1 mL of matrix
solution was mixed with 1 mL of methanolic solution of the
sample and 1 mL of the premix was loaded on the target, the
droplet was allowed to dry at ambient temperature. The
MALDI-TOF positive or negative spectra were collected in
reflectron mode.
Experimental Section
General Materials and Methods
All reagents unless stated otherwise were used as purchased
without further purification. TLC was performed on alumi-
num sheets precoated with Silica Gel 60 (F254 Merck, D);
the spots were visualized by UV light (254 nm) and/or by
spraying with 5% H2SO4 in ethanol and charred with a heat-
gun. Flash chromatography was performed with Merck silica
60 (230–400 mesh) as the stationary phase.
Enzymes
The laccase from Trametes pubescens (EC 1.10.3.2; 51 U/
mg) was kindly donated by Prof. Dietmar Haltrich (BOKU
University, Wien, Austria). The greenish lyophilisate had a
laccase activity of 51 U/mg according to the standard activi-
ty determination, by monitoring the oxidation of ABTS
[2,2’-azino-bis-(3-ethylbenzothiazoline-6-sulfonic acid)] at
436 nm. The enzyme solution (10 mL) was added to a 1-mL
NMR spectra of compounds 2 and 9 were measured in a
1
Varian Mercury 300 MHz spectrometer (299.98 MHz for H,
and 75.44 MHz for 13C), compound 4 in a Varian Unity
Inova 400 MHz spectrometer (399.87 MHz for 1H, and
100.55 MHz for 13C), compounds 3, 5, 5a, 6, 8, 12, and 13 in
2414
asc.wiley-vch.de
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 2409 – 2420

Charged Hexosaminides as New Substrates for b-N-Acetylhexosaminidase-Catalyzed Synthesis
cuvette containing 20 mM acetate buffer pH 3.5 (890 mL)
and ABTS (100 mL of a 10 mM solution of ABTS in H2O).
One enzyme unit is defined as the amount of laccase that
3.325 (m, 1H, H-5), 3.437 (m, 1H, H-4), 3.459 (m, 1H, H-5),
3.720 (ddd, 1H, J=9.9, 9.0, 8.5 Hz, H-2), 5.002 (d, 1H, J=
2.5 Hz, H-5), 5.163 (d, 1H, J=8.5 Hz, H-1), 7.171 and 8.195
(AA’BB’, 4H, SJ=9.3 Hz, pNP), 7.803 (d, 1H, J=9.0 Hz,
NH). 13C NMR (DMSO; aldehyde): d=23.05 (Ac), 54.92
(C-2), 70.41 (C-4), 73.38 (C-3), 78.94 (C-5), 98.01 (C-1),
116.69 (2ꢅC-ortho), 125.85 (2ꢅC-meta), 142.14 (C-para),
161.93 (C-ipso), 169.57 (C=O), 198.19 (C-6); 13C NMR
(DMSO; geminal diol): d=23.10 (Ac), 55.14 (C-2), 70.63
(C-4), 73.78 (C-3), 77.37 (C-5), 88.74 (C-6), 98.77 (C-1),
116.76 (2ꢅC-ortho), 125.80 (2ꢅC-meta), 141.94 (C-para),
162.35 (C-ipso), 169.49 (C=O); MS (ESI): m/z=359.0 [M+
H]+, 381.1 [M+Na]+, calcd. for C14H18N2O9: 358.1. MS con-
firmed that the aldehyde was present in the hydrated form.
p-Nitrophenyl 2-acetamido-2-deoxy-b-d-glucopyranosi-
duronic acid (3): pNP-GlcNAc (1; 300 mg, 0.876 mmol) and
(2,2,6,6-tetramethylpiperidin-1-yl)oxyl (TEMPO; 37 mg,
0.237 mmol) were dissolved in 20 mM AcONa/AcOH buffer
pH 4.5 (25 mL). Then, a solution of the laccase from Tra-
metes pubescens (765 U) in the same buffer (1 mL) was
carefully added to the suspension and the reaction was left
under gentle shaking at ambient temperature. After several
hours, the solution became clear. The reaction was moni-
tored by TLC (CHCl3: MeOH:EtOH:H2O, 6:3.5:1:0.8).
After 3.5 days, another portion of TEMPO (20 mg,
0.128 mmol) and enzyme (765 U) was added. After 5 days,
the reaction was stopped by freezing and lyophilization. Pu-
rification by gel chromatography (Biogel P2, BioRad, USA,
mobile phase water, flow rate 12.5 mLhꢀ1) afforded product
3 as a white solid; yield: 222 mg (0.623 mmol; 71%). Besides
the major product 3 (Rf 0.3), traces of 2 (Rf 0.8; <5%) were
observed in the reaction mixture; however, 2 could not be
oxidizes 1 mmol of ABTS under these conditions (eABTS
=
29.3 mMꢀ1 cmꢀ1).
Commercial plant, animal and bacterial b-N-acetylhexosa-
minidases (EC 3.2.1.52) were purchased from Sigma. The
fungal strains producing b-N-acetylhexosaminidases origi-
nated from the Culture Collection of Fungi (CCF), Depart-
ment of Botany, Charles University in Prague (CZ). The
strains were cultivated in submersed media as described pre-
viously.[4] Enzymes were obtained by (NH4)2SO4 precipita-
tion (80% saturation) of the cultivation media, and the pre-
cipitates were directly used in the enzymatic screening. The
crude enzyme preparations were tested for contaminating
sulfatase and phosphatase activities using the respective p-
nitrophenyl substrate (p-nitrophenyl sulfate, p-nitrophenyl
phosphate, Sigma) under the same conditions as the b-N-
acetylhexosaminidase assay. For the synthesis of compounds
9, 12 and 13, a substantially purified b-N-acetylhexosamini-
dase from Talaromyces flavus CCF 2686 was used. After 11
days of cultivation, the mycelium was filtered off and the
enzyme was purified from the culture medium by cation ex-
ꢀ
change chromatography [Fractogel EMD SO3 (Merck, Ger-
many), 10 mM sodium citrate/phosphate buffer pH 3.5,
linear gradient 0–1M NaCl] and gel permeation chromatog-
raphy [Superdex 200 (Amersham Biosciences, SW), 50 mM
sodium citrate/phosphate buffer pH 5 and 150 mM NaCl]
using an ꢇkta Purifier protein chromatography system
(Amersham Biosciences, SW). The b-N-acetylhexosamini-
dase activity was determined in an end-point assay as de-
scribed previously,[5] with a 2 mM starting concentration of
pNP-GlcNAc standard substrate 1 or substrates 2–6 and 8
and with 11–15 mUmLꢀ1 or 220–300 mUmLꢀ1 of b-N-ace-
tylhexosaminidase, respectively.
1
isolated using this purification method. H NMR (D2O): d=
1.806 (3H, s, 2-Ac), 3.462 (1H, dd, J=9.0, 9.6 Hz, H-4),
3.506 (1H, dd, J=9.0, 10.1 Hz, H-3), 3.752 (1H, d, J=9.6,
H-5), 3.854 (1H, dd, J=8.4, 10.1 Hz, H-2), 5.099 (1H, d, J=
8.4 Hz, H-1), 6.976 (2H, m, H-ortho), 8.026 (2H, m, H-
meta); 13C NMR (D2O): d=22.34 (2-Ac), 55.43 (C-2), 72.15
(C-4), 73.43 (C-3), 76.38 (C-5), 98.86 (C-1), 116.88 (C-
ortho), 126.35 (meta), 142.95 (C-para), 162.06 (C-ipso),
175.14 (C-6), 175.22 (2-CO); MS (ESI): found m/z=357.0
[M+H]+, 379.1 [M+Na]+, calcd. for C14H16N2O9: 356.1.
p-Nitrophenyl 2-acetamido-2-deoxy-6-O-sulfo-b-d-glu-
copyranoside, sodium salt (4): Compound 4 was prepared
as described previously,[5] and it was obtained in 52% yield
Synthesis of Substrates for b-N-Acetylhexos-
aminidases
p-Nitrophenyl 2-acetamido-2-deoxy-b-d-gluco-hexodialdo-
1,5-pyranoside (2): Dess–Martin periodinane (3.17 mL,
0.950 mmol, 0.3M solution in CH2Cl2) was added to a sus-
pension of pNP-GlcNAc (1; 300 mg, 0.876 mmol) in acetoni-
trile (3.7 mL). The mixture was stirred at room temperature
for 3 h. The reaction was stopped by adding H2O/CH2Cl2
(40 mL, 1:1 v/v) and the phases were separated. The aque-
ous phase was washed with CH2Cl2 several times, filtered,
neutralized with aqueous 1M NaOH and concentrated. The
remaining water was removed by lyophilization and the resi-
due was purified by silica gel chromatography
(CH2Cl2:MeOH, 12:1) yielding the title compound 2 as a
white amorphous solid; yield: 90 mg (0.264 mmol, 30%).
According to NMR, the sample contains two components,
the aldehyde and the geminal diol form of 2, in a varying
ratio of 34.5/65.5 (fresh solution) to 69/31 (1-day-old solu-
tion), respectively. 1H NMR (DMSO; aldehyde): d=1.812
(s, 3H, Ac), 3.460 (dd, 1H, J=10.0, 8.5 Hz, H-4), 3.577 (dd,
1, J=10.0, 8.5 Hz, H-3), 2.738 (ddd, 1H, J=10.0, 8.8, 8.2 Hz,
H-2), 4.182 (dd, 1H, J=10.0, 1.2 Hz, H-5), 5.368 (d, 1H, J=
8.2 Hz, H-1), 7.204 and 8.195 (AA’BB’, 4H, SJ=9.3 Hz,
pNP), 7.895 (d, 1H, J=8.8, 2-NH), 9.633 (d, 1H, J=1.2, H-
1
over five steps starting from pNP-GlcNAc. H NMR (D2O):
d=1.867 (3H, s, Ac), 3.459 (1H, dd, J=9.6, 9.0 Hz, H-4),
3.556 (1H, dd, J=10.2, 9.0 Hz, H-3), 3.780 (1H, ddd, J=9.6,
5.7, 2.2 Hz, H-5), 3.894 (1H, dd, J=10.2, 8.4 Hz, H-2), 4.085
(1H, dd, J=11.4, 5.7 Hz, H-6u), 4.246 (1H, dd, J=11.4,
2.2 Hz, H-6d), 5.139 (1H, d, J=8.4 Hz, H-1), 6.989 (2H,
AA’BB’, SJ=9.3 Hz, 2ꢅH-ortho), 8.019 (2H, AA’BB’, SJ=
9.3 Hz, 2ꢅH-meta); 13C NMR (D2O): d=22.38 (Ac), 55.48
(C-2), 67.16 (C-6), 69.68 (C-4), 73.51 (C-3), 74.38 (C-5),
98.85 (C-1), 116.76 (2ꢅC-ortho), 126.36 (2ꢅC-meta), 142.87
(C-para), 161.88 (C-ipso), 175.19 (C=O). MS (ESI): found
m/z=422.7
C14H18N2O11S: m/z=422.0.
[M+H]+,
444.7
[M+Na]+,calcd.
for
p-Nitrophenyl 2-acetamido-2-deoxy-3,4-di-O-acetyl-6-
O-phosphate-b-d-glucopyranoside (5a): POCl3 (0.5 mL,
5.462 mmol) was added dropwise under an argon atmos-
phere to a solution of dry p-nitrophenyl 2-acetamido-2-
1
6); H NMR (DMSO; geminal diol): d=1.804 (s, 3H, Ac),
Adv. Synth. Catal. 2011, 353, 2409 – 2420
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2415

Pavla Bojarovꢀ et al.
FULL PAPERS
deoxy-3,4-di-O-acetyl-b-d-glucopyranoside[6] (1a; 300 mg,
0.703 mmol) and dry ethyldiisopropylamine (DIPEA) in
CH2Cl2 (11 mL, 1:10 v/v). After 15 h of stirring at ambient
temperature, another portion of DIPEA (5 mL) was added,
followed by the addition of water (1 mL). The reaction mix-
ture was stirred for a further 1 h and, after the addition of
EtOH (20 mL), it was evaporated to dryness. After purifica-
tion by silica gel chromatography (mobile phase CHCl3:
MeOH, 10:3), compound 5a was obtained as a white solid;
p-Nitrophenyl 2-acetamido-2-deoxy-6-O-phospho-b-d-
galactopyranoside, cyclohexylamine salt (8): The synthesis
of compound 8 (galacto-) is analogous to that of 5 ACHTUNGTRENNUNG(gluco-).
POCl3 (0.4 mL, 4.370 mmol) was added dropwise under an
argon atmosphere to the solution of dry p-nitrophenyl 2-
acetamido-2-deoxy-3,4-di-O-acetyl-b-d-galactopyranoside[6]
(7; 200 mg, 0.469 mmol) and dry DIPEA in CH2Cl2 (11 mL,
1:10 v/v). After 15 h of stirring at ambient temperature, an-
other portion of DIPEA (5 mL) was added, followed by the
addition of water (1 mL). The reaction mixture was stirred
for a further 1 h and, after the addition of EtOH (20 mL), it
was evaporated to dryness. The resulting yellow oil was dis-
solved in MeOH (20 mL) and DOWEX 50W resin (Sigma,
5 g) in H+ form (prewashed with MeOH) was added. After
1 min of stirring, the reaction mixture was filtered and
evaporated to dryness under vacuum. The yellowish solid
was dissolved in dry methanol (5 mL) and a solution of
sodium methoxide in methanol (25% w/w) was added drop-
wise, until the pH of the reaction mixture reached 13. After
the starting material disappeared (TLC; acetonitrile:water,
4:1), the reaction mixture was directly loaded onto Sepha-
dex LH-20 (Sigma) column (mobile phase MeOH, flow rate
15 mLhꢀ1). Fractions containing the product were collected
and after evaporation mixed with DOWEX 50W resin in
methanol. The resin was filtered off and after adding cyclo-
hexylamine (5 mL), the mixture was dried under vacuum.
The product was crystallized from water/i-PrOH to give 8 as
white crystals; yield: 115 mg (0.272 mmol, 58%). 1H NMR
(D2O): d=0.974 (1H, m, H-4’d), 1.097 (1H, m, H-3’d), 1.138
(1H, m, H-2’u), 1.437 (1H, m, H-4’u), 1.587 (1H, m, H-3’u),
1.778 (1H, m, H-2’u), 1.800 (3H, s, 2-Ac), 2.944 (1H, m, H-
1’), 3.697 (1H, dd, J=3.3, 10.9 Hz, H-3), 3.808 (2H, m, H-
6), 3.865 (1H, m, H-5), 3.925 (1H, br d, J=3.3 Hz, H-4),
4.025 (1H, dd, J=8.4, 10.9 Hz, H-2), 5.061 (1H, d, J=8.4,
H-1), 7.023 (2H, m, H-ortho), 8.053 (2H, m, H-meta);
13C NMR (D2O): d=22.41 (2-Ac), 24.05 (C-3’), 24.55 (C-4’),
1
yield: 300 mg (0.591 mmol, 84%). H NMR (DMSO-d6): d=
1.769 (3H, s, 2-Ac), 1.941 (3H, s, 3-Ac), 1.990 (3H, s, 4-Ac),
3.740 (1H, ddD, J=6.0, 11.6 Hz, JH,P =8.4 Hz, H-6u), 3.843
(1H, ddD, J=1.9, 11.6 Hz, JH,P =6.1 Hz, H-6d), 4.047 (1H,
dd, J=8.4, 10.2 Hz, H-2), 4.106 (1H, ddd, J=1.9, 6.0,
9.8 Hz, H-5), 4.860 (1H, dd, J=9.6, 9.8 Hz, H-4), 5.213 (1H,
dd, J=9.6, 10.2 Hz, H-3), 5.493 (1H, d, J=8.4, H-1), 7.247
(2H, m, H-ortho), 8.106 (1H, d, J=9.0 Hz, 2-NH), 8.196
(2H, m, H-meta), 9.692 [2H, -OP(=O)(OH)2]; 13C NMR
(DMSO-d6): d=20.41 (3-Ac), 20.49 (4-Ac), 22.69 (2-Ac),
53.08 (C-2), 63.08 (C-6, JC,P =4 Hz), 68.90 (C-4), 72.66 (C-3),
72.70 (C-5), 97.44 (C-1), 116.86 (C-ortho), 125.82 (C-meta),
142.21 (C-para), 161.69 (C-ipso), 169.28 (4-CO), 169.50 (2-
CO), 169.72 (3-CO); 31P NMR: d=ꢀ1.581.
p-Nitrophenyl 2-acetamido-2-deoxy-6-O-phospho-b-d-
glucopyranoside, sodium salt (5): A solution of sodium
methoxide in methanol (25% w/w) was added dropwise to
the suspension of 5a (250 mg, 0.434 mmol) in dry methanol
(5 mL), until the pH of the reaction mixture reached 13.
After complete disappearance of the starting material as ob-
served by TLC (acetonitrile:water, 4:1), the reaction mix-
ture was directly loaded onto a Sephadex LH-20 (Sigma)
column (mobile phase MeOH, flow rate 15 mLhꢀ1). Com-
pound 5 was obtained as a white solid; yield: 100 mg
1
(0.208 mmol, 48%). H NMR (D2O): d=1.803 (3H, s, 2-Ac),
3.503 (1H, dd, J=9.1, 10.0 Hz, H-3), 3.525 (1H, ddd, J=1.9,
3.6, 9.2 Hz, H-5), 3.577 (1H, dd, J=9.1, 9.2 Hz, H-4), 3.824
(1H, ddD, J=1.9, 12.2 Hz, JH,P =5.9 Hz, H-6u), 3.852 (1H,
dd, J=8.4, 10.0 Hz, H-2), 3.892 (1H, ddD, J=3.6, 12.2 Hz,
30.59 (C-2’), 50.66 (C-1’), 52.45 (C-2), 63.07 (C-6, JC,P
=
4 Hz), 67.32 (C-4), 70.68 (C-3), 74.55 (C-5, JC,P =7 Hz), 99.49
(C-1), 116.87 (C-ortho), 126.42 (C-meta), 142.91 (C-para),
162.16 (C-ipso), 175.34 (2-CO); 31P NMR (D2O): d=1.910.
MS (ESI): m/z=445.0 [M+Na]+, 466.9 [M+Na2]+, calcd.
for C14H19N2O11P: 422.1.
J
H,P =7.7 Hz, H-6d), 5.115 (1H, d, J=8.4, H-1), 7.006 (2H,
m, H-ortho), 8.044 (2H, m, H-meta); 13C NMR (D2O): d=
22.37 (2-Ac), 55.82 (C-2), 62.64 (C-6, JC,P =4 Hz), 69.33 (C-
4), 73.12 (C-3), 76.23 (C-5, JC,P =7 Hz), 99.09 (C-1), 116.85
(C-ortho), 126.41 (C-meta), 142.96 (C-para), 162.02 (C-ipso),
175.15 (2-CO); 31P NMR: d=4.759. MS (ESI): found m/z=
467.0 [M+Na2]+: calcd. for C14H19N2O11P: 422.1.
Molecular Modeling
Molecular dynamics simulation was performed using the
p-Nitrophenyl 2-acetamido-2-deoxy-6-O-sulfo-b-d-gal-
actopyranoside, sodium salt (6): Compound 6 was synthe-
sized analogously to compound 4;[5] it was obtained in a
previously reported dimeric models of the b-N-acetylhexos-
AHCTUNGTRENNUNG
aminidases from Aspergillus oryzae[19] and Penicillium oxali-
cum.[18] All substrates were built with Yasara, and optimized
by AM1 semi-empirical method.[24] Force field parameters
were assigned using the AutoSMILES approach.[25] Initial
ligand positions for docking experiments were determined
by overlaying the substrate and enzyme with the crystal
structure of bacterial chitobiase cocrystallized with the
ligand (pdb code 1QBB) in Yasara.[26] In the second step the
actual docking was performed using the version of Auto-
dock 4.0[27] implemented in Yasara applying local Search Al-
gorithm, grid space 0.375 nm, to determine the exact docked
position of the standard substrate pNP-GlcNAc (1) as well
as of individual C-6-modified substrates 2–5 in the active
site. For molecular docking, the pyranosyl rings of the sub-
strates were in a slightly distorted 1,4B/1S3 conformation to
1
30% overall yield. H NMR (DMSO-d6): d=1.795 (3H, s, 2-
Ac), 3.597 (1H, m, H-3), 3.732 (1H, ddd, SJ=7.3 Hz, H-4),
3.819 (1H, dd, J=12.0, 8.5 Hz, H-6u), 3.912 (2H, m, H-5,
H-6d), 4.038 (1H, ddd, J=10.6, 9.0, 8.4 Hz, H-2), 4.79 (1H,
br s, 3-OH), 4.805 (1H, d, J=4.5 Hz, 4-OH), 5.132 (1H, d,
J=8.4 Hz, H-1), 7.185 (2H, AA’BB’, SJ=9.3 Hz, H-ortho),
7.743 (1H, d, J=9.0 Hz, 2-NH), 8.188 (2H, AA’BB’, SJ=
9.3 Hz, H-meta); 13C NMR (DMSO): d=23.08 (2-Ac), 51.65
(C-2), 65.03 (C-6), 67.54 (C-4), 70.88 (C-3), 73.47 (C-5),
98.91 (C-1), 116.70 (C-ortho), 125.75 (C-meta), 141.82 (C-
para), 162.38 (C-ipso), 169.62 (CO); MS (ESI): found m/z=
422.7 [M+H]+, 444.7 [M+Na]+, calcd. for C14H18N2O11S:
422.0.
2416
asc.wiley-vch.de
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 2409 – 2420

Charged Hexosaminides as New Substrates for b-N-Acetylhexosaminidase-Catalyzed Synthesis
resemble the ꢈreactiveꢆ conformations in the Michaelis struc-
tures reported by Vocadlo and co-workers.[28]
Synthesis of Oligosaccharides by b-N-Acetylhexos-
aminidases
The molecular dynamics simulations of the substrate-
enzyme complexes were run to further improve the docked
positions, including the solvation effect, and to assess the
stability of the docked position in the active site. All simula-
tions were run for 3 ns in explicit TIP3[29] water with Yasara
using the Yamber 2[30] force field; periodic boundary condi-
tions with the simulation cell extended 0.8 nm on each side
of the complex. Sodium ions were iteratively placed at the
coordinates with the lowest electrostatic potential until the
cell was neutral. Intramolecular forces were calculated every
1 fs, intermolecular every 2 fs; Lennard-Jones and electro-
static interactions were cut off at 0.76 nm, long-range elec-
trostatic interactions were computed using the Particle-
Mesh Ewald method,[31] with grid spacing 0.1 nm, fourth-
order B-splines, and a tolerance of 10ꢀ4 for the direct space
sum. The simulation was run in the following NPT ensem-
ble: constant temperature (298 K), pressure and number of
particles. The evaluation of ligand-enzyme complexes in
time was analyzed on the basis of geometry and energy pa-
rameters. Root mean square deviation and binding energies
were calculated with Yasara. Binding energies were calculat-
ed using a method applied earlier for substrate-galactosami-
nidase complexes.[32] It considers the internal energy ob-
tained with the specified force field, including van der Waals
solvation energy and a correction for the entropic cost of ex-
posing the substrate/enzyme surface to water.[33,34] The solva-
tion energy was calculated using the boundary fast method
implemented in Yasara. Free energies of binding were calcu-
lated by AutoDock with Lamarckian Genetic Algorithm
and grid space 0.285 nm[17] using substrate-enzyme com-
plexes minimized in Yasara after 3 ns of MD.
p-Nitrophenyl 2-acetamido-2-deoxy-b-d-gluco-pyranosylur-
onic acid-(1!4)-2-acetamido-2-deoxy-b-d-gluco-pyranosi-
duronic acid (9): Substrate 3 (100 mg, 0.281 mmol) was dis-
solved in sodium citrate/phosphate buffer pH 5 (936 mL).
The b-N-acetylhexosaminidase from Talaromyces flavus
CCF 2686 (30 U) was added and the mixture was shaken at
358C. After 3.5 h, the reaction was stopped by heating to
1008C for 2 min. The reaction mixture was cooled to room
temperature and centrifuged (13,000 rpm, 10 min), concen-
trated under vacuum and loaded onto a Biogel P2 (BioRad,
USA) column (water, flow rate 12.5 mLhꢀ1). Fractions con-
taining product 9 were collected, lyophilized and further pu-
rified by reversed-phase HPLC. HPLC was carried out in a
modular system consisting of DeltaChrom SDS 020 and 030
pumps (Watrex, CZ), a Spectra 100 Variable UV/VIS CE
detector (Thermo Separation Products, USA), Basic Mara-
thon Plus autosampler (Watrex, CZ) and dyn./stat. mixing
chamber (SunChrom GmbH, D). Clarity AS software
(Chromservis, CZ) was used for evaluation. A Luna (C8)
column, 250ꢅ4.6 mm, with a 4ꢅ3 mm guard column (Phe-
nomenex, USA), was used at ambient temperature for ana-
lytical scale analysis (mobile phase MeCN:H2O 20:80, flow
rate 0.8 mLminꢀ1). For the purification of product 9 a Bios-
pher SI C8 5 mm semipreparative column 250ꢅ8 mm
(Watrex, CZ) was used (mobile phase MeCN:H2O 20:80,
flow rate 3.2 mLminꢀ1). The retention time of compound 9
was 5.89 min (analytical column) and 3.97 min (preparative
column). Compounds were detected at 210 nm. Disaccharide
9 was obtained as a white solid; yield: 7.7 mg (0.022 mmol,
1
16%). H NMR (D2O): d=1.719 (3H, s, 2’-Ac), 1.742 (3H,
s, 2-Ac), 3.338 (2H, m, H-3’,H-4’), 3.492 (1H, dd, J=8.4,
10.3 Hz, H-2’), 3.595 (1H, dd, J=8.5, 10.0 Hz, H-3), 3.679
(1H, dd, J=8.5, 9.2 Hz, H-4), 3.766 (1H, d, J=9.7 Hz, H-5),
3.815 (1H, dd, J=8.2, 10.0 Hz, H-2), 3.977 (1H, d, J=
9.2 Hz, H-5), 4.377 (1H, d, J=8.4 Hz, H-1’), 5.102 (1H, d,
J=8.2 Hz, H-1), 6.857 (2H, m, H-ortho), 7.911 (2H, m, H-
meta); 13C NMR (D2O, HSQC and HMBC readouts): d=
22.4 (2’-Ac), 22.7 (2-Ac), 54.7 (C-2), 55.4 (C-2’), 71.7 (C-3,
C-3ꢆ or C-4’), 73.2 (C-3ꢆ or C-4’), 74.0 (C-5), 74.5 (C-5’), 80.1
(C-4), 98.8 (C-1), 101.6 (C-1’), 116.8 (C-ortho), 126.5 (C-
meta), 143.1 (C-para), 161.6 (C-ipso), 170.9 (C-6), 172.4 (C-
6’), 175.0 (2’-CO), 175.1 (2-CO); MS (MALDI-TOF): m/z=
574.2 [M+H]+, 596.2 [M+Na]+: calcd. for C22H27N3O15:
573.1.
Analytical Transglycosylation Reactions
For autocondensation reactions with substrates 3–5 (where
the substrate served both as glycosyl donor and acceptor),
the respective substrate (25–300 mM) was dissolved in
sodium citrate/phosphate buffer and the b-N-acetylhexosa-
minidase from T. flavus CCF 2686 (2.5–30 UmLꢀ1) was
added. For the glycosylation of GlcNAc with substrates 2, 4
and 5, the respective substrate (50–150 mM) and 2-acetami-
do-2-deoxy-d-glucopyranose (11; 200–600 mM) were dis-
solved in sodium citrate/phosphate buffer and the b-N-ace-
tylhexosaminidase from T. flavus CCF 2686 (2.5–30 UmLꢀ1)
was added. Phosphate 5 was also tested as a glycosyl accept-
or: pNP-GlcNAc (50 mM) and 5 (200 mM) were dissolved
in sodium citrate/phosphate buffer and the b-N-acetylhexo-
saminidase from T. flavus CCF 2686 (1.8 UmLꢀ1) was
added. For the reactions with substrates 2--4, 50 mM sodium
citrate/phosphate buffer pH 5 was used. With phosphate
substrate 5, 200 mM sodium citrate/phosphate buffer pH 4
was used and after the substrate dissolved, the pH was ad-
justed to 5 with 200 mM citric acid. Here, a stronger buffer-
ing capacity was required to suppress the alkalization of the
reaction medium by phosphate 5. The reactions were incu-
bated at 358C with shaking (850 rpm) for 24 h. Aliquots
were taken at regular time intervals and analyzed by TLC
(2-propanol:H2O:NH3 aq., 7: 2: 1).
2-Acetamido-2-deoxy-b-d-gluco-pyranosyluronic
acid-
(1!4)-2-acetamido-2-deoxy-d-gluco-pyranose (10): Com-
pound 2 (51 mg, 0.150 mmol) and 2-acetamido-2-deoxy-d-
glucopyranose (11; 232 mg, 1.050 mmol) were suspended in
50 mM sodium citrate/phosphate buffer pH 5 (2.96 mL), the
b-N-acetylhexosaminidase from Talaromyces flavus CCF
2686 [(NH4)2SO4 precipitate; 6 U] was added and the reac-
tion was incubated at 378C with shaking (850 rpm). The re-
action progress was monitored by TLC (2-propanol:
H2O:NH4OH aq., 7:2:1). After 4 h, the reaction was stopped
by boiling for 2 min. The reaction mixture was cooled to
room temperature and NaClO2 (38 mg, 0.337 mmol) was
added. After 5 h, the oxidation was complete, the reaction
mixture was centrifuged (13,500 rpm, 10 min), concentrated
under vacuum and loaded onto Bio Gel P2 (BioRad, USA)
Adv. Synth. Catal. 2011, 353, 2409 – 2420
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2417

Pavla Bojarovꢀ et al.
FULL PAPERS
column (mobile phase water, flow rate 9.5 mLhꢀ1). Disac-
charide 10 was obtained as a white solid; yield: 15 mg
(0.034 mmol; yield 23% referred to donor 2). According to
50 mm sodium citrate/phosphate buffer pH 5 (1200 mL), the
b-N-acetylhexosaminidase from Talaromyces flavus CCF
2686 (1 U) was added and the reaction was incubated at
358C with shaking (850 rpm). After 6.5 h, the reaction was
stopped by boiling for 2 min. The reaction mixture was
cooled to room temperature and centrifuged (13,000 rpm,
10 min), concentrated under vacuum and loaded onto a
Biogel P2 (BioRad, USA) column (water, flow rate
11.5 mLhꢀ1). Disaccharide 13 was obtained as a white solid;
yield: 16 mg (0.030 mmol, 33%). According to NMR, 13 was
NMR, 10 was
a mixture of two anomers (a/b=1.6).
1H NMR (D2O; a-anomer): d=1.855 (3H, s, 2-Ac), 1.888
(3H, s, 2’-Ac), 3.389 (1H, m, H-4’), 3.416 (1H, ddd, J=10.0,
8.5 Hz, H-4), 3.421 (1H, m, H-3’), 3.483 (1H, m, H-6u),
3.587 (1H, m, H-5’), 3.599 (1H, m, H-6d), 3.607 (1H, m, H-
2’), 3.667 (1H, dd, J=10.8, 3.5 Hz, H-2), 3.704 (1H, ddd, J=
10.0, 5.0, 2.0 Hz, H-5), 3.732 (1H, dd, J=10.8, 8.5 Hz, H-3),
4.415 (1H, d, J=8.5 Hz, H-1’), 5.018 (1H, d, J=3.5 Hz, H-
1); 1H NMR (D2O; b-anomer): d=1.855 (3H, s, 2-Ac),
1.888 (3H, s, 2’-Ac), 3.342 (1H, ddd, J=9.8, 5.5, 2.0 Hz, H-
5), 3.382 (1H, m, H-4’), 3.405 (1H, m, H-4), 3.414 (1H, m,
H-3’), 3.462 (1H, m, H-6u), 3.502 (1H, m, H-2), 3.538 (1H,
m, H-3), 3.578 (1H, m, H-5’), 3.594 (1H, m, H-2’), 3.640
(1H, m, H-6d), 4.412 (1H, d, J=8.5 Hz, H-1’), 4.515 (1H, d,
J=8.4 Hz, H-1); 13C NMR (D2O; a-anomer): d=22.16 (2-
Ac), 22.39 (2’-Ac), 53.92 (C-2), 55.72 (C-2’), 60.39 (C-6),
69.47 (C-3), 70.18 (C-5), 72.18 (C-4’), 73.63 (C-3’), 75.68 (C-
5’), 80.45 (C-4), 90.62 (C-1), 101.49 (C-1’), 174.71 (CO),
174.85 (COꢆ), 175.46 (C-6’; 13C NMR (D2O; b-anomer): d=
22.16 (2-Ac), 22.39 (2’-Ac), 55.70 (C-2’), 56.35 (C-2), 60.53
(C-6), 72.18 (C-4’), 72.76 (C-3), 73.61 (C-3’), 74.81 (C-5),
75.72 (C-5’), 80.04 (C-4), 95.12 (C-1), 101.46 (C-1’), 174.78
(COꢆ), 175.01 (CO), 175.46 (C-6’). MS (MALDI-TOF):
m/z=439.06 [M+H]+, 461.18 [M+Na]+, 477.15 [M+K]+,
calcd. for C16H26N2O12: 438.15.
1
a mixture of two anomers (a/b=1.8). H NMR (DMSO-d6;
a-anomer): d 1.820, 1.822 (2ꢅ3H, 2ꢅs, 2-Ac, 2’-Ac), 3.105
(1H, dd, SJ=16.5 Hz, H-4’), 3.285 (1H, dd, SJ=17.8 Hz, H-
4), 3.32 (1H, m, H-3’), 3.35 (1H, m, H-5’), 3.458 (1H, m, H-
2’), 3.47 (1H, m, H-6u), 3.56 (1H, m, H-6d), 3.572 (2H, m,
H-2, H-5), 3.663 (1H, dd, SJ=18.2 Hz, H-3), 3.781 (1H, dd,
J=12.2, 6.6 Hz, H-6’u), 4.070 (1H, m, H-6’d), 4.367 (1H, d,
J=8.2 Hz, H-1’), 4.932 (1H, d, J=3.2 Hz, H-1), 7.729 (1H,
br d, 2’-NH), 7.782 (1H, d, J=7.5 Hz, 2-NH); 1H NMR
(DMSO-d6; b-anomer): d=1.820, 1.822 (2ꢅ3H, 2ꢅs, 2-Ac,
2’-Ac), 3.105 (1H, dd, SJ=16.5 Hz, H-4’), 3.162 (1H, m, H-
5), 3.261 (1H, dd, SJ=17.8 Hz, H-4), 3.32 (1H, m, H-3’),
3.35 (1H, m, H-5’), 3.36 (1H, m, H-2), 3.374 (1H, m, H-6u),
3.441 (1H, m, H-2’), 3.508 (1H, dd, SJ=18.0 Hz, H-3), 3.614
(1H, m, H-6d), 3.781 (1H, dd, J=12.2, 6.6 Hz, H-6’u), 4.070
(1H, m, H-6’d), 4.387 (1H, d, J=8.0 Hz, H-1’), 4.462 (1H,
d, J=8.3 Hz, H-1), 7.729 (1H, br d, 2’-NH), 7.812 (1H, d,
J=8.4 Hz, 2-NH); 13C NMR (DMSO-d6; a-anomer): d=
22.67, 23.03 (2-Ac, 2’-Ac), 53.88 (C-2), 55.47 (C-2’), 60.05
(C-6), 65.55 (C-6’), 68.64 (C-3), 69.88 (C-5), 70.30 (C-4’),
73.62 (C-3’), 74.98 (C-5’), 81.70 (C-4), 90.12 (C-1), 102.05
(C-1’), 169.11 (COꢆ), 169.31 (CO); 13C NMR (DMSO-d6; b-
anomer): d=22.67, 23.03 (2-Ac, 2’-Ac), 55.47 (C-2’), 56.65
(C-2), 60.24 (C-6), 65.61 (C-6’), 70.35 (C-4’), 72.23 (C-3),
73.71 (C-3’), 74.85 (C-5), 74.98 (C-5’), 81.28 (C-4), 95.45 (C-
1), 101.83 (C-1’), 169.15 (COꢆ), 169.75 (CO); MS (ESI):
m/z=504.7 [M+H]+, 526.7 [M+Na]+, calcd. for
C16H28N2O14S: 504.1.
p-Nitrophenyl
2-acetamido-2-deoxy-6-O-sulfo-b-d-
gluco-pyranosyl-(1!4)-2-acetamido-2-deoxy-6-O-sulfo-b-
d-gluco-pyranoside (12): Substrate 4 (60 mg, 0.135 mmol)
was dissolved in sodium citrate/phosphate buffer pH 5
(1350 mL). The b-N-acetylhexosaminidase from Talaromyces
flavus CCF 2686 (0.2 U) was added and the mixture was
shaken at 358C. After 4 h, the reaction was stopped by heat-
ing to 1008C for 2 min. The reaction mixture was cooled to
room temperature and centrifuged (13,000 rpm, 10 min),
concentrated under vacuum and loaded onto a Biogel P2
(BioRad, USA) column (water, flow rate 11.5 mLhꢀ1). Dis-
accharide 12 was obtained as a white solid; yield: 14 mg
Competitive Inhibition Assay with NKR-P1 and
CD69 Activation Receptors of NK Cells
1
(0.019 mmol, 28%). H NMR (DMSO-d6): d=1.820 (3H, s,
2-Ac), 1.895 (3H, s, 2’-Ac), 3.130 (2H, m, H-3’, H-4’), 3.386
(1H, m, H-5’), 3.511 (1H, dd, SJ=18.2 Hz, H-4), 3.608 (1H,
m, H-2’), 3.640 (1H, dd, SJ=18.3 Hz, H-3), 3.770 (1H, ddd,
J=10.2, 8.7, 8.4 Hz, H-2), 3.812 (1H, dd, J=11.3, 6.1 Hz, H-
6’u), 3.837 (1H, m, H-5), 3.896 (1H, dd, J=11.1, 1.9 Hz, H-
6u), 4.064 (1H, dd, J=11.1, 3.9 Hz, H-6d), 4.077 (1H, dd,
J=11.3, 2.0 Hz, H-6’d), 4.436 (1H, d, J=8.6 Hz, H-1’), 5.279
(1H, d, J=8.4 Hz, H-1), 7.217 (2H, AA’BB’, SJ=9.3 Hz, H-
ortho), 7.665 (1H, d, J=9.5 Hz, 2’-NH), 8.023 (1H, d, J=
8.7 Hz, 2-NH), 8.173 (2H, AA’BB’, SJ=9.3 Hz, H-meta);
13C NMR (DMSO-d6): d=22.96 (2’-Ac), 22.99 (2-Ac), 54.68
(C-2’), 54.74 (C-2), 63.70 (C-6), 65.52 (C-6’), 70.23 (C-4’),
71.76 (C-3), 72.53 (C-5), 74.69 (C-3’), 75.15 (C-5’), 80.55 (C-
4), 97.72 (C-1), 101.95 (C-1’), 116.72 (C-ortho), 125.72 (C-
meta), 141.97 (C-para), 162.02 (C-ipso), 169.29 (COꢆ), 169.67
(CO); MS (ESI): m/z=705.7 [M+H]+, 727.7 [M+Na]+,
calcd. for C22H31N3O19S2: 705.1.
The inhibition assays were performed as described previous-
ly,[35] with the following modification: soluble NKR-P1 and
CD69 protein receptors were labeled with fluorescent labels
(fluorescein and rhodamine, respectively). The concentra-
tions of bound protein receptors in the microtiter wells were
determined by fluorescence measurement (lex/lem =496/
519 nm and lex/lem =546/577 nm for NKR-P1 and CD69, re-
spectively) using a Safire 2 spectrophotometer (Tecan, AT).
The results are given as a negative logarithm of the ligand
concentration required to cause 50% inhibition of the recep-
tor binding to the standard high-affinity ligand
GlcNAc23BSA (ꢀlog IC50). Proteins were labeled by cova-
lently attaching fluorescent labels using N-hydroxysuccini-
mide fluorescein and N-hydroxysuccinimide rhodamine
(both by Pierce Biotechnology, USA) for rat NKR-P1A and
human CD69 receptors, respectively.
2-Acetamido-2-deoxy-6-O-sulfo-b-d-gluco-pyranosyl-
(1!4)-2-acetamido-2-deoxy-d-gluco-pyranose (13): Com-
pound 4 (40 mg, 0.090 mmol) and 2-acetamido-2-deoxy-d-
gluco-pyranose (11; 78 mg, 0.353 mmol) were suspended in
The 96-well round-bottomed plate was coated with
GlcNAc23BSA ligand, blocked with 2% BSA and after incu-
bation at 48C for 2 h, the plate was washed three times with
PBS. The labeled proteins and serial dilutions of studied
2418
asc.wiley-vch.de
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 2409 – 2420

Charged Hexosaminides as New Substrates for b-N-Acetylhexosaminidase-Catalyzed Synthesis
compounds (1–6, 8–13 and respective reference compounds, References
see Table 2) were put into each well, incubated at 48C for
1 h, then the plate was washed three times with PBS and in-
cubated at 48C overnight with 0.1M sodium acetate buffer
supplemented with 0.1% octyl b-d-gluco-pyranoside and
0.1% Triton X-100. Then, the solution was transferred to 96-
well flat-bottomed UV-transparent plates and the results
were obtained by fluorescence measurement. Complete in-
hibition curves were constructed and the IC50 values were
calculated from at least three independent experiments.
ˇ
[1] K. Slꢀmovꢀ, P. Bojarovꢀ, L. Petrꢀskovꢀ, V. Kren, Bio-
technol. Adv. 2010, 28, 682–693.
[2] a) S. Singh, J. Packwood, C. J. Samuel, P. Critchley,
D. H. G. Crout, Carbohydr. Res. 1995, 279, 293–305;
b) S. Singh, M. Scigelova, P. Critchley, D. H. G. Crout,
Carbohydr. Res. 1997, 305, 363–370; c) L. Weignerovꢀ,
ˇ
ˇ
ˇ
P. Vavruskovꢀ, A. Pisvejcovꢀ, J. Thiem, V. Kren, Carbo-
hydr. Res. 2003, 338, 1003–1008; d) N. Aboitiz, F. J.
ˇ
ˇ
CaÇada, L. Husꢀkovꢀ, M. Kuzma, V. Kren, J. Jimꢃnez-
Barbero, Org. Biomol. Chem. 2004, 2, 1987–1994.
[3] a) B. L. Mark, D. J. Mahuran, M. M. Cherney, D. Zhao,
S. Knapp, M. N. G. James, J. Mol. Biol. 2003, 327, 1093–
1109; b) M. J. Lemieux, B. L. Mark, M. M. Cherney,
S. G. Withers, D. J. Mahuran, M. N. G. James, J. Mol.
Biol. 2006, 359, 913–929.
Immunological Tests
Rat NK cells were purified from mononuclear spleen cells
using Ficoll-Paque centrifugation (Sigma) followed by the
immunodepletion of T cells, B cells, and monocytes, and
subsequent IL-2 activation in order to enhance the surface
expression of the NKR-P1A receptor.[36] Peripheral blood
mononuclear cells were obtained from standard blood frac-
tions enriched in leukocytes (buffy coats from the local
Blood Transfusion service) after dilution with RPMI1640
medium (Sigma), and Ficoll-Paque centrifugation. Cells
were incubated overnight in complete RPMI1640 medium
in plastic cell culture dishes to allow the adherent cells to
attach. Collected non-adherent fraction of PBMC (N-
PBMC) contained mostly lymphocytes (T, B, and NK cells).
Lymphocytes from donors expressing CD69 in less than 5%
of cells were designated CD69low. Lymphocytes from donors
with more than 20% CD69-positive cells were further acti-
vated by incubation at a density of 2ꢀ106 cellsmLꢀ1 in com-
plete RPMI1640 medium for 4 h with PMA (50 ngmLꢀ1)
and ionomycin (500 ngmLꢀ1). This procedure increased the
surface expression of CD69 to 75–85%, as analyzed by flow
cytometry using monoclonal antibody against CD69 labeled
with phycoerythrin. Such lymphocytes were designated as
CD69high. Cellular activation assays and tests of natural kill-
ing were essentially performed as described previously.[9b]
For apoptosis assays, the cells were resuspended at a density
of 2ꢀ106 mLꢀ1 in complete RPMI1640 medium, aliquoted
into round-bottomed 96-well plates, and the tested com-
pounds in various concentrations were added into duplicate
wells. The compounds were added 12 and 6 h before deter-
mining the percentage of apoptotic cells using Annexin V-
FITC/Hoechst 33258 staining and flow cytometry. The per-
centage of apoptotic cells (Annexin V+/Hoechst 33258ꢀ) ob-
served in the presence of PBS alone and in the presence of
5ꢀ10ꢀ6 M arsenite were used as the negative and positive
control, respectively.
ˇ
[4] P. Fialovꢀ, L. Weignerovꢀ, J. Rauvolfovꢀ, V. Prikrylovꢀ,
ˇ
A. Pisvejcovꢀ, R. Ettrich, M. Kuzma, P. Sedmera, V.
Kren, Tetrahedron 2004, 60, 693–701.
[5] K. J. Loft, P. Bojarovꢀ, K. Slꢀmovꢀ, V. Kren, S. J. Wil-
liams, ChemBioChem 2009, 10, 565–576.
[6] K. Slꢀmovꢀ, R. Gazꢀk, P. Bojarovꢀ, N. Kulik, R. Et-
trich, H. Pelantovꢀ, P. Sedmera, V. Kren, Glycobiology
2010, 20, 1002–1009.
[7] P. Fialovꢀ, A. T. Carmona, I. Robina, R. Ettrich, P. Sed-
mera, V. Prikrylovꢀ, L. Petrꢀskovꢀ-Husꢀkovꢀ, V. Kren,
Tetrahedron Lett. 2005, 46, 8715–8718.
[8] K. Slꢀmovꢀ, P. Marhol, K. Bezouska, L. Lindkvist, S. G.
ˇ
ˇ
ˇ
ˇ
ˇ
ˇ
ˇ
ˇ
ˇ
Hansen, V. Kren, H. H. Jensen, Bioorg. Med. Chem.
Lett. 2010, 20, 4263–4265.
[9] a) P. Krist, E. Herkommerovꢀ-Rajnochovꢀ, J. Rauvol-
ˇ
ˇ
ˇ
fovꢀ, T. Semenuk, P. Vavruskovꢀ, J. Pavlꢂcek, K. Be-
ˇ
ˇ
ˇ
zouska, L. Petrus, V. Kren, Biochem. Biophys. Res.
Commun. 2001, 287, 11–20; b) A. Kovalovꢀ, M. Ledvi-
na, D. Saman, D. Zyka, M. Kubꢂckovꢀ, L. Zꢂdek, V.
Sklenꢀr, P. Pompach, D. Kavan, J. Bꢂly, O. Vanek, Z.
Kubꢂnkovꢀ, M. Libigerovꢀ, L. Ivanovꢀ, M. Antolꢂkovꢀ,
ˇ
ˇ
ˇ
ˇ
´
ˇ
´
ˇ
H. Mrꢀzek, D. Rozbesky, K. Hofbauerovꢀ, V. Kren, K.
Bezouska, J. Med. Chem. 2010, 53, 4050–4065.
ˇ
ˇ
[10] K. Bezouska, C. Yuen, J. O’Brien, R. A. Childs, W.
ˇ
ˇ
Chai, A. M. Lawson, K. Drbal, A. Fiserovꢀ, M. Pospꢂsil,
T. Feizi, Nature 1994, 372, 150–157.
[11] P. Fialovꢀ, D.-J. Namdjou, R. Ettrich, V. Prikrylovꢀ, J.
ˇ
ˇ
Rauvolfovꢀ, K. Krenek, M. Kuzma, L. Elling, K. Be-
zouska, V. Kren, Adv. Synth. Catal. 2005, 347, 997–
ˇ
ˇ
1006.
ˇ
[12] P. Bojarovꢀ, K. Krenek, M. Kuzma, L. Petrꢀskovꢀ, K.
ˇ
ˇ
Bezouska, D.-J. Namdjou, L. Elling, V. Kren, J. Mol.
Catal. B: Enzymatic 2008, 50, 69–73.
Acknowledgements
ˇ
ˇ
ˇ
ˇ
[13] K. Bezouska, R. Snajdrovꢀ, K. Krenek, M. Vancurovꢀ,
A. Kꢀdek, D. Adꢀmek, P. Lhotꢀk, D. Kavan, K. Hof-
Financial support by the Czech Science Foundation (grants
203/09/P024 to P.B.; 305/09/H008; 303/09/0477 to K.B.;
P207/11/0629 to V. K.), by the research concepts
AV0Z50200510, AV0Z60870520 and MSM21620808, and by
the Ministry of Education of the Czech Republic (LC06010)
is gratefully acknowledged.
ˇ
bauerovꢀ, P. Man, P. Bojarovꢀ, V. Kren, Bioorg. Med.
Chem. 2010, 18, 1434–1440.
[14] J. Neumann, S. Weingarten, J. Thiem, Eur. J. Org.
Chem. 2007, 1130–1144.
[15] P. Sedmera, J. Volc, V. Havlꢂcek, S. Pakhomova, A. Je-
ˇ
gorov, Carbohydr. Res. 1997, 297, 375–378.
[16] S. Freimund, S. Kopper, Carbohydr. Res. 2004, 339,
217–220.
Adv. Synth. Catal. 2011, 353, 2409 – 2420
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2419

Pavla Bojarovꢀ et al.
FULL PAPERS
[17] G. M. Morris, D. S. Goodsell, R. S. Halliday, R. Huey,
W. E. Hart, R. K. Belew, A. J. Olson. J. Comp. Chem.
1998, 19, 1639–1662.
[27] D. S. Goodsell, A. J. Olson, Proteins – Struct. Funct. Bi-
oinformat. 1990, 8, 195–202.
[28] Y. He, M. S. Macauley, K. A Stubbs, D. J Vocadlo, G. J.
Davies, J. Am. Chem. Soc. 2010 132, 1807–1809.
[29] W. L. Jorgensen, J. Tirado-Rives, J. Am. Chem. Soc.
1988, 110, 1657–1666.
ˇ
ˇ
[18] H. Ryslavꢀ, A. Kalendovꢀ, V. Doubnerovꢀ, P. Skocdo-
ˇ
ˇ
pol, V. Kumar, Z. Kukacka, P. Pompach, O. Vanek, K.
ˇ
Slꢀmovꢀ, P. Bojarovꢀ, N. Kulik, R. Ettrich, V. Kren, K.
ˇ
Bezouska, FEBS J. 2011, 278, 2469–2484.
[30] E. Krieger, T. Darden, S. B. Nabuurs, A. Finkelstein, G.
´
[19] R. Ettrich, V. Kopecky Jr, K. Hofbauerovꢀ, V. Baum-
Vriend, Proteins 2004, 57, 678–683.
´
ruk, P. Novꢀk, P. Pompach, P. Man, O. Plꢂhal, M. Kuty,
[31] U. Essman, L. Perera, M. L. Berkowitz, T. Darden, H.
Lee, L. G. Pedersen, J. Chem. Phys. B 1995, 103, 8577–
8593.
ˇ
ˇ
ˇ
ˇ
N. Kulik, J. Sklenꢀr, H. Ryslavꢀ, V. Kren, K. Bezouska,
BMC Struct. Biol. 2007, 7, 32.
[20] M. Ogata, X. Zeng, T. Usui, H. Uzawa, Carbohydr.
[32] N. Kulik, L. Weignerovꢀ, T. Filipi, P. Pompach, P.
Res. 2007, 342, 23–30.
ˇ
ˇ
Novꢀk, H. Mrꢀzek, K. Slꢀmovꢀ, K. Bezouska, V. Kren,
ˇ
[21] P. Bojarovꢀ, K. Krenek, K. Wetjen, K. Adamiak, H. Pe-
R. Ettrich, Glycobiology 2010, 20, 1410–1419.
ˇ
ˇ
lantovꢀ, K. Bezouska, L. Elling, V. Kren, Glycobiology
2009, 19, 509–517.
[33] P. Bultinck. Computational medicinal chemistry for
drug discovery, 1st edn, CRC Press, Boca Raton, 2004.
[34] A. Leach. Molecular modelling: principles and applica-
tions, 2nd edn., Dorchester, Prentice Hall, 2001.
ˇ
[22] O. Renaudet, K. Krenek, I. Bossu, P. Dumy, A. Kꢀdek,
ˇ
ˇ
ˇ
D. Adꢀmek, O. Vanek, D. Kavan, R. Gazꢀk, M. Sulc,
ˇ
ˇ
K. Bezouska, V. Kren, J. Am. Chem. Soc. 2010, 132,
ˇ
´
[35] a) J. Pavlꢂcek, B. Sopko, E. R. Kopecky, V. Baumruk, P.
6800–6808.
ˇ
´
Man, V. Havlꢂcek, M. Vrbacky, L. Martꢂnkovꢀ, V.
ˇ
[23] C. T. Yuen, K. Bezouska, J. OꢆBrien, M. Stoll, R. Le-
ˇ
ˇ
ˇ
Kren, M. Pospꢂsil, K. Bezouska, Biochemistry 2003, 42,
moine, A. Lubineau, M. Kiso, A. Hasegawa, N. J. Bock-
ovich, K. C. Nicolaou, T. Feizi, J. Biol. Chem. 1994,
269, 1595–1598.
ˇ
9295–9306; b) O. Vanek, M. Nꢀlezkovꢀ, D. Kavan, I.
ˇ
Borovickovꢀ, P. Pompach, P. Novꢀk, V. Kumar, L. Van-
ˇ
´
nucci, J. Hudecek, K. Hofbauerovꢀ, V. Kopecky, J.
[24] M. J. S. Dewar, E. G. Zoebisch, E. F. Healy, J. J. P. Stew-
art, J. Am. Chem. Soc. 1985, 107, 3902–3909.
[25] A. Jakalian, D. B. Jack, C. I. Bayly, J. Comput. Chem.
2002, 23, 1623–1641.
[26] I. Tews, A. Perrakis, A. Oppenheimer, Z. Dauter, K. S.
Wilson, C. E. Vorgias, Nat. Struct. Biol. 1996, 3, 638–
648.
ˇ
Brynda, P. Kolenko, J. Dohnꢀlek, P. Kaderꢀvek, J.
ˇ
ˇ
ˇ
ˇ
Chmelꢂk, L. Gorcꢂk, L. Zꢂdek, V. Sklenꢀr, K. Bezouska,
FEBS J. 2008, 275, 5589–5606.
ˇ
[36] K. Bezouska, Biochem. Soc. Trans. 1996, 24, 156–161.
2420
asc.wiley-vch.de
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 2409 – 2420