Enzymatic synthesis of new C-6-acylated derivatives of NAG-thiazoline and evaluation of their inhibitor activities towards fungal β-N- acetylhexosaminidase

Article abstract of DOI:10.1016/j.molcatb.2012.10.016

β-N-Acetylhexosaminidases (EC 3.2.1.52) from the CAZy glycoside hydrolase families 20 and 84 are two distinct enzyme groups with similar reactivity and different physiological functions, thus selective inhibition of these enzymes is of crucial importance.

Full text of DOI:10.1016/j.molcatb.2012.10.016

Journal of Molecular Catalysis B: Enzymatic 87 (2013) 128–134
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
Enzymatic synthesis of new C-6-acylated derivatives of NAG-thiazoline and
evaluation of their inhibitor activities towards fungal ␤-N-acetylhexosaminidase
a
a
a
a
b
Jana Krejzováa,c, Petr Simon , Eva Vavrˇíková , Kristy´na Slámová , Helena Pelantová , Sergio Riva ,
ˇ
Vojteˇch Spiwokc, Vladimír Krˇena,∗
a Institute of Microbiology, Academy of Sciences of the Czech Republic, Vídenˇská 1083, CZ-142 20 Prague 4, Czech Republic
b Istituto di Chimica del Riconoscimento Molecolare, C.N.R., Via Mario Bianco 9, I-20131 Milano, Italy
c Department of Biochemistry and Microbiology, Institute of Chemical Technology Prague, Technická 5, CZ-166 28 Prague 6, Czech Republic
a r t i c l e i n f o
a b s t r a c t
Article history:
␤-N-Acetylhexosaminidases (EC 3.2.1.52) from the CAZy glycoside hydrolase families 20 and 84 are two
distinct enzyme groups with similar reactivity and different physiological functions, thus selective inhi-
bition of these enzymes is of crucial importance. Here, we report on the lipase-catalyzed synthesis of a set
of novel monomeric and dimeric C-6-acylated derivatives of NAG-thiazoline, which is a typical competi-
tive inhibitor of both these enzyme classes. The prepared compounds were tested as potential inhibitors
of a fungal GH20 ␤-N-acetylhexosaminidase from Talaromyces flavus, however, the results of the inhi-
bition tests were quite ambiguous. The observed inhibition was generally weak with some features of
competitive inhibition (increase of KM), but the overall pattern appeared indecisive.
Received 12 September 2012
Received in revised form 25 October 2012
Accepted 25 October 2012
Available online 16 November 2012
Keywords:
NAG-thiazoline
Lipase
© 2012 Elsevier B.V. All rights reserved.
Vinyl ester
␤-N-Acetylhexosaminidase
Inhibitor
1. Introduction
benzyl ␣-d-mannopyranoside [7]. The importance of regioselecti-
vity of lipase-catalyzed reactions is reflected also in the chemistry
Biocatalysis in organic chemistry has become
a
standard
of flavonoids. Optically pure pentaacetylated silybins were selec-
tively deacetylated by lipase AK in good conversion rate (40%).
Two products – 3,20,23-tri-O-acetyl-silybin and 3,5,20,23-tetra-
O-acetyl-silybin – were obtained in high yields applicable for the
subsequent synthetic processes [8].
practice. Generally, enzymatic routes are characterized by mild
reaction conditions and high selectivity. Lipases (EC 3.1.1.3), largely
used in organic synthesis, naturally catalyze the hydrolysis of
water-insoluble esters such as triglycerides. They act effectively in
organic solvents or at the water/organic interface and lipases are
frequently used for the preparation of enantio-, chemo- or regiose-
lectively pure esters(transesterification) and carboxylic acids (ester
hydrolysis). Moreover, lipase-catalyzed protection/deprotection of
reactive groups is often suitable and useful in multi-step organic
synthesis [1,2].
Regioselective acylation of carbohydrates has been applied in
many biocatalytic studies. Recently this methodology has been
exploited for the transesterification of saccharides with long
chain fatty acids as biodegradable emulsifiers [3–5] and regios-
elective acylation at 5ꢀ-OH of the monosaccharide prodrug of
5-fluorouridin [6]. Moreover, Novozym 435 (lipase from Candida
antarctica immobilized on acrylic resin) has been successfully
applied in the two-step regioselective synthesis of C-6 hybrid
modified glycosides, e.g. phenyl ␤-d-glucopyranoside linked to
␤-N-Acetylhexosaminidases (EC 3.2.1.52) are glycoside hydro-
lases catalyzing the cleavage of terminal ␤-d-GlcNAc and
␤-d-GalNAc residues in various oligosaccharides and glycocon-
jugates. ␤-N-Acetylhexosaminidase from the filamentous fungus
Talaromyces flavus CCF 2686 has been deeply studied from the view-
point of its unusually broad substrate specificity and great transgly-
cosylation abilities [9–11], but has never been investigated in inhi-
bition assays. Due to importance of this enzyme for biotechnolog-
ical applications its detailed kinetical as well as inhibitory param-
eters should be known. Besides it, inhibitory parameters of fungal
hexosaminidases in general have been studied rather scarcely [12].
The most detailed information on the eukaryotic hex-
osaminidases was acquired on the human ones. Generally,
two unrelated classes of enzymes able to cleave GlcNAc-moieties
from various biomolecules were found in the human body: isoen-
zymes of lysosomal GH family 20 ␤-N-acetylhexosaminidases
HexA and HexB [13–15], and nucleocytoplasmic ␤-N-
acetylglucosaminidase belonging to the glycoside hydrolase family
84 [16–18]. The function, biochemical properties and structure of
∗
Corresponding author. Tel.: +420 296 442 510; fax: +420 296 442 509.
E-mail address: kren@biomed.cas.cz (V. Krˇen).
1381-1177/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molcatb.2012.10.016

J. Krejzová et al. / Journal of Molecular Catalysis B: Enzymatic 87 (2013) 128–134
129
␤-N-acetylhexosaminidases from humans have been subjects of
intensive research, as both types of enzymes are important in
different physiological processes and their dysfunction or absence
causes severe neurodegenerative disorders [19–21]. Moreover,
both classes share the substrate-assisted catalytic mechanism [22];
and also the architecture of their active site is similar [23]. Thus, the
search for the inhibitors specific for one of the two enzyme groups
is a challenging task, as the highly selective inhibitors are neces-
sary tools for the study of the functions of the respective enzymes
in vivo, as well as for the development of therapeutics with minimal
side effects [22,24]. A wide spectrum of small molecule inhibitors
of the enzymes from the GH families 20 and 84 has been developed
and tested [25]. The general strategy is to design a compound
mimicking the transition state with a special focus on the selective
inhibition of just one of the two enzyme groups.
influence the inhibition. Moreover, recently a multivalency effect
of the inhibitors of glycosidases was demonstrated in fullerens dec-
orated with imino-sugar-type inhibitors [30]. Subsequently, the
prepared compounds were tested as potential inhibitors of ␤-N-
acetylhexosaminidase from the filamentous fungus Talaromyces
flavus. Even though the parent compound NAG-thiazoline (1) has
been referred to as a potent competitive inhibitor of plant and
human lysosomal GH20 ␤-N-acetylhexosaminidases and human
GH84 O-GlcNAcase [22,26], with the fungal enzyme used in
this study the observed inhibition pattern was not clearly com-
petitive. Moreover, the C-6 acylated NAG-thiazoline derivatives
displayed weak inhibition properties that could be identified as
non-competitive (7), un-competitive (10) and also mixed (9, 14,
15) inhibition types using Lineweaver–Burk plots.
The oxazoline transition state analogue NAG-thiazoline (1;
Fig. 1) was designed to prove the substrate-assisted catalytic
mechanism of GH20 ␤-N-acetylhexosaminidases, showing a good
inhibition potential with these enzymes [26,27]. Unfortunately,
NAG-thiazoline was also found to inhibit GH84 O-GlcNAcase in
a similar manner, which means that this inhibitor does not dis-
criminate between both target enzyme groups, human lysosomal
␤-N-acetylhexosaminidases and O-GlcNAcase [22]. To investigate
the possibility to increase selectivity of the NAG-thiazoline scaf-
fold, a series of NAG-thiazoline derivatives with larger moieties
at the 2ꢀ-position of the thiazoline ring were synthesized and
tested with both enzymes, showing up to 3100-fold inhibition
selectivity with O-GlcNAcase over human HexB [22]. Subsequently,
after structure optimization, Thiamet-G, the highly selective NAG-
thiazoline-based inhibitor was designed and tested, affording a
37,000-fold selectivity for O-GlcNAcase over HexB, which makes
this compound useful for the study of O-GlcNAcylation in vivo
[24]. Some more functionalized NAG-thiazolines have been pre-
pared, namely azide and fluoride derivatives showing outstanding
selectivity for O-GlcNAcase, which may find use as reagents for
imaging the O-GlcNAc cycle in living cells and tissues [28]. The C-1-
homologated thiazolines exhibited significantly reduced inhibition
against selected GH20 enzymes relative to NAG-thiazoline [29].
In this study, we present the lipase-catalyzed synthesis of a
series of novel C-6 acylated NAG-thiazoline derivatives (Fig. 1)
– monomers and dimers – using the respective vinyl esters as
activated acyl donors. The concept of the dimeric inhibitors was
designed due to the fact that fungal hexosaminidases are dimers of
two functional subunits [12] and, therefore, multivalency effect can
2. Experimental
2.1. Materials
Chemicals were obtained from Sigma–Aldrich. Reactions were
monitored and the purity of prepared products was checked
by TLC (Merck silica gel/TLC sheets 60 F254). The lipase from
Candida antarctica immobilized on acrylic resin (Novozym 435)
was purchased from Sigma–Aldrich. Lipase CE5 was obtained from
Amano. NAG-thiazoline (2ꢀ-methyl-␣-d-glucopyrano-[2,1-d]-ꢀ2ꢀ-
thiazoline, 1) was prepared according to the literature procedure
via the reaction of Lawesson’s reagent with the peracetylated
␤-anomer of N-acetylglucosamine followed by Zemplén deacety-
lation [26].
2.2. General methods
NMR spectra were measured on a Bruker AVANCE III 400 MHz
spectrometer (400.13 MHz for 1H, and 100.61 MHz for 13C) in
CD3OD at 303 K. Residual signal of solvent was used as an inter-
nal standard (ıH 3.330, ıc 49.30). 1H NMR, 13C NMR, COSY, HSQC,
and HMBC spectra were measured using standard manufacturers’
software. Chemical shifts are given in ı-scale [ppm], and coupling
constants in Hz. Digital resolution enabled us to report chemical
shifts of protons to three and carbon chemical shifts to two decimal
places. Some hydrogen chemical shifts were read out from HSQC
and are reported to two decimal places. Proton spin systems were
assigned by COSY, then transferred to carbons by HSQC and joined
together using heteronuclear correlations in HMBC. The thiazoline
Fig. 1. Structures of NAG-thiazoline (1) and of its derivatives prepared and tested in this study.

130
J. Krejzová et al. / Journal of Molecular Catalysis B: Enzymatic 87 (2013) 128–134
structure was proved by the doublet of methyl (J = 2.0 Hz) correlated
to carbons C-1 and C-2. Formation of divinyl ester 15 was confirmed
by the auto-correlation cross peak of C-2ꢀ, the respective cross peak
for divinyl ester 14 was not unambiguously distinguished.
Mass spectra were measured on MALDI-TOF/TOF ultraFLEX III
mass spectrometers (Bruker–Daltonics, Bremen, Germany). Pos-
itive spectra were calibrated externally using the monoisotopic
[M+H]+ ions of PepMixII calibrant (Bruker–Daltonics, Bremen). For
MALDI experiment 0.4 ␮l of sample dissolved in 50% acetonitrile
was allowed to dry at ambient temperature on the target and over-
laid with matrix solution. The positive or negative spectra were
collected in a reflectron mode. The high resolution EI spectra were
measured using a GCT Premier benchtop orthogonal acceleration
time-of-flight mass spectrometer (Waters, Milford, MA, USA). The
samples were loaded into a quartz cup of direct insertion probe and
introduced into the electron ionization source (200 ◦C). Standard
70 eV spectra were internally calibrated with perfluorotributy-
lamine (PFTBA). Data were acquired and processed using MassLynx
4.1 (Waters) software.
2.3.2. General procedure A – preparation of C-6-modified
NAG-thiazolines (7–11)
NAG-thiazoline 1 (219 mg, 1 mmol) was suspended in anhy-
drous acetone (15 ml). Vinyl esters of the acid (2b–6b) (2 mmol,
2 equiv.), sodium carbonate (2.8 equiv., 300 mg) and 300 mg of
Novozym 435 were added to the mixture. The reaction mixture
was incubated at 45 ◦C, 500 rpm in a Thermomixer (Eppendorf,
DE) for 16 h. Reaction mixture was evaporated in vacuo and the
product was purified by flash chromatography on silica gel (100%
EtOAc).
((3aR,5R,6S,7R,7aR)-6,7-Dihydroxy-2-methyl-5,6,7,7a-
tetrahydro-3aH-pyrano[3,2-d]thiazol-5-yl)methyl
acetate
(7):
white powder (yield 82%), 1H NMR (400.13 MHz, CD3OD): 2.070
(3H, s, Ac), 2.284 (3H, d, J = 2.0 Hz, CH3) 3.518 (1H, dddd, J = 9.2,
6.4, 2.6, 0.5 Hz, H-5), 3.613 (1H, ddd, J = 9.2, 3.8, 1.0 Hz, H-4),
4.158 (1H, ddd, J = 4.8, 3.8, 0.5 Hz, H-3), 4.355 (1H, dddq, 7.0, 4.8,
1.0, 2.0 Hz, H-2), 4.207 (1H, dd, J = 12.0, 6.4 Hz, H-6u), 4.274 (1H,
dd, J = 12.0, 2.6 Hz, H-6d), 6.345 (1H, d, J = 6.97.0 Hz, H-1). 13C
NMR (100.61 MHz, CD3OD): 20.94 (CH3), 20.99 (Ac), 65.56 (C-6),
71.73 (C-4), 73.92 (C-5), 74.31 (C-3), 80.84 (C-2), 90.68 (C-1),
2.3. Chemistry
170.95 ( C
), 173.01 (6-CO). HRMS m/z [M+H]+ calc. 262.07492,
measured 262.07422.
2.3.1. Preparation of vinylesters (3b–6b) – general procedure
((3aR,5R,6S,7R,7aR)-6,7-Dihydroxy-2-methyl-5,6,7,7a-
Acid (3a–6a, 10 mmol, 1 equiv.) was dissolved in dry THF (1 ml)
together with vinyl acetate (110 mmol, 11 equiv.), then palla-
dium(II) acetate (22 mg, 0.1 mmol, 0.01 equiv.) was added and the
mixture was stirred at 60 ◦C for 16 h. The reaction mixture was
cooled down to 20 ◦C and evaporated in vacuo. The obtained oil
was purified by flash chromatography on silica gel (EtOAc:hexane
3:1) yielding title compounds in the first fraction as a yellowish oil.
Vinyl 3-phenylpropanoate (3b): yellowish oil (yield 87%). Pre-
viously, compound 3b was published [31]. 1H NMR (400.13 MHz,
CD3OD): 2.762 (2H, m, COCH2), 3.037 (2H, m, ꢁJ = 15.5 Hz, CH2),
4.618 (1H, dd, J = 6.3, 1.6 Hz, CH2u), 4.929 (1H, dd, J = 14.0, 1.6 Hz,
CH2d), 7.265 (3H, m, H-ortho, H-para), 7.332 (1H, dd, J = 14.0,
6.3 Hz, CH), 7.346 (2H, m, H-meta). 13C NMR (100.61 MHz, CD3OD):
30.54 (CH2), 35.49 (COCH2), 98.97 ( CH2), 126.35 (C-para), 128.23
tetrahydro-3aH-pyrano[3,2-d]thiazol-5-yl)methyl
3-phenylpropanoate (8): white powder (yield 87%), 1H NMR
(400.13 MHz, CD3OD): 2.251 (3H, d, J = 2.0 Hz, CH3), 2.646 (2H, m,
H-2ꢀ), 2.922 (2H, t, J = 7.6 Hz, H-1ꢀ), 3.533 (1H, ddd, J = 9.2, 6.5, 2.6 Hz,
H-5), 3.615 (1H, ddd, J = 9.2, 4.0, 0.9 Hz, H-4), 4.161 (1H, dd, J = 5.1,
4.0 Hz, H-3), 4.224 (1H, dd, J = 12.0, 6.5 Hz, H-6u), 4.295 (1H, dd,
J = 12.0, 2.6 Hz, H-6d), 4.338 (1H, dddq, J = 6.9, 5.1, 0.9, 2.0 Hz, H-2),
6.309 (1H, d, J = 6.9 Hz, H-1), 7.197 (1H, m, H-para), 7.205 (2H, m,
H-ortho), 7.271 (2H, m, H-meta). 13C NMR (100.61 MHz, CD3OD):
20.98 (CH3), 32.04 (C-2ꢀ), 36.86 (C-1ꢀ), 65.45 (C-6), 71.65 (C-4),
73.96 (C-5), 74.28 (C-3), 80.77 (C-2), 90.63 (C-1), 127.49 (C-para),
129.54 (C-ortho), 129.73 (C-meta), 142.05 (C-ipso), 170.88 (
C ),
174.62 (6-CO). HRMS m/z [M+H]+ calc. 352.12187, measured
352.12112.
(C-ortho), 128.52 (C-meta), 140.06 (C-ipso), 141.12 (
C
), 169.87
((3aR,5R,6S,7R,7aR)-6,7-Dihydroxy-2-methyl-5,6,7,7a-
(CO). HRMS m/z [M+H]+ calc. 176.0837, measured 176.0840.
Vinyl 2-phenylacetate (4b): yellowish oil (yield 38%). Previously,
compound 4b was published [32]. 1H NMR (400.13 MHz, CD3OD):
3.712 (2H, s, CH2), 4.597 (1H, dd, J = 6.3, 1.6 Hz, CH2u), 4.896 (1H,
dd, J = 14.0, 1.6 Hz, CH2d), 7.265 (1H, m, H-para), 7.276 (1H, dd,
J = 14.0, 6.3 Hz, CH), 7.287 (2H, m, H-ortho), 7.323 (2H, m, H-meta).
13C NMR (100.61 MHz, CD3OD): 41.70 (CH2), 98.61 ( CH2), 128.49
(C-para), 129.85 (C-meta), 130.63 (C-ortho), 135.13 (C-ipso), 142.78
tetrahydro-3aH-pyrano[3,2-d]thiazol-5-yl)methyl 2-phenylacetate
(9): white powder (yield 75%), 1H NMR (400.13 MHz, CD3OD):
2.248 (3H, d, J = 2.0 Hz, CH3), 3.511 (1H, dddd, J = 9.2, 6.5, 2.4,
0.5 Hz, H-5), 3.596 (1H, ddd, J = 9.2, 4.1, 0.9 Hz, H-4), 3.669 (2H,
s, CH2), 4.135 (1H, ddd, J = 5.0, 4.1, 0.5 Hz, H-3), 4.244 (1H, dd,
J = 12.0, 6.5 Hz, H-6u), 4.299 (1H, dddq, J = 6.9, 5.0, 0.9, 2.0 Hz, H-2),
4.325 (1H, dd, J = 12.0, 2.4 Hz, H-6d), 6.232 (1H, d, J = 6.9 Hz, H-1),
7.26 (1H, m, H-para), 7.31 (2H, m, H-ortho), 7.32 (2H, m, H-meta).
13C NMR (100.61 MHz, CD3OD): 20.91 (CH3), 42.11 (CH2), 65.85
(C-6), 71.57 (C-4), 74.18 (C-5), 74.29 (C-3), 81.07 (C-2), 90.48 (C-1),
128.32 (C-para), 129.80 (C-meta), 130.67 (C-ortho), 135.71 (C-ipso),
(
CH ), 170.58 (CO). HRMS m/z [M+H]+ calc. 162.0681, measured
162.0680.
Vinyl 2-phenoxyacetate (5b): yellowish oil (yield 72%). Previ-
ously, compound 5b was published [33]. 1H NMR (400.13 MHz,
CD3OD): 4.710 (1H, dd, J = 6.2, 1.6 Hz, CH2u), 4.729 (2H, s, OCH2),
5.010 (1H, dd, J = 13.9, 1.6 Hz, CH2d), 6.960 (2H, m, H-ortho), 7.043
(1H, m, H-para), 7.332 (2H, m, H-meta), 7.360 (1H, dd, J = 13.9,
6.2 Hz, CH). 13C NMR (100.61 MHz, CD3OD): 64.86 (OCH2), 98.97
170.82 ( C
), 173.47 (6-CO). HRMS m/z [M+H]+ calc. 338.10622,
measured 338.10559.
((3aR,5R,6S,7R,7aR)-6,7-Dihydroxy-2-methyl-5,6,7,7a-
tetrahydro-3aH-pyrano[3,2-d]thiazol-5-yl)methyl 2-phenoxyacetate
(10): white powder (yield 70%), 1H NMR (400.13 MHz, CD3OD):
2.257 (3H, d, J = 2.0 Hz, CH3) 3.560 (1H, ddd, J = 9.2, 6.1, 2.8 Hz,
H-5), 3.625 (1H, ddd, J = 9.2, 4.0, 0.9 Hz, H-4), 4.155 (1H, dd, J = 5.1,
4.0 Hz, H-3), 4.321 (1H, dddq, J = 6.9, 5.1, 0.9, 2.0 Hz, H-2), 4.361
(1H, dd, J = 12.0, 6.1 Hz, H-6u), 4.412 (1H, dd, J = 12.0, 2.8 Hz, H-6d),
4.718 (2H, s, CH2), 6.295 (1H, d, J = 6.9 Hz, H-1), 6.931 (2H, m,
H-ortho), 6.982 (1H, m, H-para), 7.292 (2H, m, H-meta). 13C NMR
(100.61 MHz, CD3OD): 20.93 (CH3), 65.96 (C-6), 66.22 (CH2), 71.52
(C-4), 73.98 (C-5), 74.24 (C-3), 80.87 (C-2), 90.51 (C-1), 115.95
(C-ortho), 122.88 (C-para), 130.82 (C-meta), 159.52 (C-ipso), 170.96
(
(
CH2), 114.67 (C-ortho), 121.92 (C-para), 129.54 (C-meta), 140.49
CH ), 157.56 (C-ipso), 166.18 (CO). HRMS m/z [M+H]+ calc.
178.0630, measured 178.0626.
Vinyl 2-(4-hydroxyphenyl)acetate (6b): yellowish oil (yield 14%),
1H NMR (400.13 MHz, CD3OD): 3.609 (2H, s, CH2), 4.595 (1H, dd,
J = 6.3, 1.6 Hz, CH2u), 4.892 (1H, dd, J = 14.0, 1.6 Hz, CH2d), 6.773
(2H, m, ꢁJ = 8.7 Hz, H-meta), 7.107 (2H, m, ꢁJ = 8.7 Hz, H-ortho),
7.257 (1H, dd, J = 14.0, 6.3 Hz, CH). 13C NMR (100.61 MHz, CD3OD):
40.86 (CH2), 98.62 ( CH2), 116.65 (C-meta), 125.86 (C-ipso), 131.67
(C-ortho), 142.77 ( CH ), 157.84 (C-para), 171.22 (CO). HRMS m/z
[M+H]+ calc. 178.0630, measured 178.0636.
(
C
), 171.08 (6-CO). HRMS m/z [M]+ calc. 353.09331, measured
353.26628.

J. Krejzová et al. / Journal of Molecular Catalysis B: Enzymatic 87 (2013) 128–134
131
((3aR,5R,6S,7R,7aR)-6,7-Dihydroxy-2-methyl-5,6,7,7a-
tetrahydro-3aH-pyrano[3,2-d]thiazol-5-yl)methyl
2-(4-hydroxyphenyl)acetate (11): white powder (yield 76%),
1H NMR (400.13 MHz, CD3OD): 2.261 (3H, d, J = 2.1 Hz, CH3), 3.482
(1H, dddd, J = 9.2, 6.5, 2.6, 0.5 Hz, H-5), 3.566 (2H, s, CH2), 3.586
(1H, ddd, J = 9.2, 4.0, 0.9 Hz, H-4), 4.123 (1H, ddd, J = 4.9, 4.0, 0.5 Hz,
H-3), 4.215 (1H, dd, J = 12.0, 6.5 Hz, H-6u), 4.298 (1H, dddq, J = 6.9,
5.1, 0.9, 2.1 Hz, H-2), 4.306 (1H, dd, J = 12.0, 2.6 Hz, H-6d), 6.225
(1H, d, J = 6.9 Hz, H-1), 6.758 (2H, m, ꢁJ = 8.7 Hz, H-meta), 7.116
(2H, m, ꢁJ = 8.7 Hz, H-ortho). 13C NMR (100.61 MHz, CD3OD): 20.84
(CH3), 41.35 (CH2), 65.83 (C-6), 71.65 (C-4), 74.25 (C-5), 74.33
(C-3), 81.18 (C-2), 90.48 (C-1), 116.63 (C-meta), 126.55 (C-ipso),
Scheme 1. Preparation of vinyl esters (3b–6b).
131.72 (C-ortho), 157.80 (C-para), 170.99 (
C ), 174.10 (6-CO).
[10]. The inhibition tests were carried out by measuring the
enzyme activity at a constant substrate concentration (0.25 mM)
with the variation of the concentration of the respective inhibitors
(0.01–0.5 mM) under the above conditions. Michaelis–Menten
kinetics of ␤-N-acetylhexosaminidase reactions was measured in
a kinetic assay at 348 nm (isosbestic point of p-nitrophenol), using
pNP-GlcNAc as substrate with its concentration in the range of
0.04–0.5 mM (citrate-phosphate buffer pH 7, 25 ◦C); in the inhibi-
tion assays the concentration of the respective inhibitors ranged
from 0.001 to 2 mM. All data were measured in five parallels.
HRMS m/z [M]+ calc. 353.09331, measured 353.26625.
2.3.3. General procedure B – preparation of diesters of
NAG-thiazoline (14, 15)
Divinyl esters of dicarboxylic acids (12, 13) were synthe-
sized according to the published procedure [7]. NAG-thiazoline
(1, 1.37 mmol, 1 equiv.) and divinyl ester of dioic acid (3.42 mmol,
2.5 equiv.) were dissolved in anhydrous acetone (15 ml). 300 mg of
Novozym 435 and 50 mg of molecular sieves 4A were added to the
solution. The reaction mixture was incubated at 45 ◦C, 200 rpm in a
Thermomixer (Eppendorf, DE). After 36 h, the reaction was stopped
by filtration of the enzyme and the solvent was evaporated under
reduced pressure. The crude product was purified by silica gel flash
chromatography (EtOAc:MeOH 95:5).
3. Results and discussion
3.1. Synthesis of the C-6 acylated derivatives of NAG-thiazoline
bis(((3aR,5R,6S,7R,7aR)-6,7-Dihydroxy-2-methyl-5,6,7,7a-
tetrahydro-3aH-pyrano[3,2-d]thiazol-5-yl)methyl) dodecanedioate
(14): compound 14 was prepared according to general procedure
B as a white solid (120 mg, 28%). 1H NMR (400.13 MHz, CD3OD):
1.341 (12H, m, H-4ꢀ, H-5ꢀ, H-6ꢀ), 1.636 (4H, m, H-3ꢀ), 2.284 (6H, d,
J = 2.0 Hz, CH3), 2.357 (4H, t, J = 7.4 Hz, H-2ꢀ), 3.511 (2H, dddd, J = 9.2,
6.7, 2.5, 0.5 Hz, H-5), 3.604 (2H, ddd, J = 9.2, 3.7, 1.0 Hz, H-4), 4.179
(2H, dd, J = 4.7, 3.7 Hz, H-3), 4.217 (2H, dd, J = 12.0, 6.8 Hz, H-6u),
4.291 (2H, dd, J = 12.0, 2.5, H-6d), 4.362 (2H, dddq, J = 7.0, 4.7, 1.0,
2.0 Hz, H-2), 6.339 (2H, d, J = 7.0 Hz, H-1). 13C NMR (100.61 MHz,
CD3OD): 20.95 (CH3), 26.33 (C-3ꢀ) 30.38 (C-4ꢀ), 30.61 (C-5ꢀ), 30.76
(C-6ꢀ), 35.28 (C-2ꢀ), 65.45 (C-6), 71.88 (C-4), 74.02 (C-5), 74.29
This study deals with lipase-catalyzed regioselective forma-
tion of C-6 esters and dimers of the ␤-N-acetylhexosaminidase
inhibitor NAG-thiazoline and the testing of the inhibition activi-
ties of the prepared compounds. Generally, the preparation of the
C-6 derivatives of NAG-thiazoline comprised two steps: synthesis
of the activated acyl donors and lipase-mediated acylation of NAG-
thiazoline, yielding the series of the target compounds. We have
opted for the use of the vinyl esters of acids, dicarboxylic acids
and acids substituted with aromatic rings because it is known that
the subsequent lipase-catalyzed reactions are faster and the reac-
tions equilibria are more shifted in favour of acylation than in the
reactions where free acids are employed.
The syntheses of the vinyl esters 3b–6b (Scheme 1) was carried
out by the modified method of Piri [35]. The target esters of sub-
stituted acids were obtained from the reactions with vinyl acetate
under the Pd(II) catalysis with THF as a solvent at 60 ◦C. Synthe-
ses of these products, e.g. vinyl 3-phenylpropanoate (3b), vinyl
2-phenylacetate (4b), vinyl 2-phenoxyacetate (5b) had already
been described [31–33], whereas the synthesis of vinyl 2-(4-
hydroxyphenyl)acetate (6b) has not been published yet.
Lipase-mediated synthesis is a preferable and very effective
method for selective acylation of primary hydroxyl groups in non-
aqueous media. Here we used Novozym 435 – lipase from Candida
antarctica immobilized on an acrylic resin – taking advantage from
its specific selectivity and facile handling. Thus, five C-6 acylated
derivatives of NAG-thiazoline (7–11; Scheme 2) were synthesized
under the catalysis of lipase Novozym 435 in dry acetone, in which
NAG-thiazoline was completely soluble and the lipase did not lose
its biocatalytic activity. The thiazoline ring is unstable under even
mild acidic conditions; therefore the addition of sodium carbon-
ate was necessary as a prevention of the undesired hydrolysis.The
divinyl esters of dicarboxylic acids (12, 13) were synthesized fol-
lowing a different way. Aliphatic dicarboxylic acids were converted
to their divinyl esters in the presence of vinyl acetate and a catalytic
amount of mercury(II) acetate in accordance with Magrone et al.
[7]. In the lipase-catalyzed reactions two units of NAG-thiazoline
were linked by the short (15) or long (14) aliphatic chain. Gener-
ally, the design of divalent and multivalent molecules was shown
(C-3), 80.94 (C-2), 90.68 (C-1), 170.99 (
C
), 175.77 (C-1ꢀ). HRMS
m/z [M+Na]+ calc. 655.23351, measured 655.23242.
bis(((3aR,5R,6S,7R,7aR)-6,7-Dihydroxy-2-methyl-5,6,7,7a-
tetrahydro-3aH-pyrano[3,2-d]thiazol-5-yl)methyl)
pentanedioate
(15): compound (15) was prepared according to general procedure
B as a white solid (196 mg, 53%). 1H NMR (400.13 MHz, CD3OD):
1.937 (2H, qui, J = 7.3 Hz, H-3ꢀ), 2.284 (6H, d, J = 2.0 Hz, CH3), 2.446
(4H, t, J = 7.3 Hz, H-2ꢀ), 3.520 (2H, dddd, J = 9.2, 6.6, 2.6, 0.4 Hz, H-5),
3.613 (2H, ddd, J = 9.2, 3.8, 0.9 Hz, H-4), 4.164 (2H, ddd, J = 4.8, 3.8,
0.4 Hz, H-3), 4.235 (2H, dd, J = 12.0, 6.6 Hz, H-6u), 4.312 (2H, dd,
J = 12.0, 2.6, H-6d), 4.355 (2H, dddq, J = 7.0, 4.8, 0.9, 2.0 Hz, H-2),
6.346 (2H, d, J = 7.0 Hz, H-1). 13C NMR (100.61 MHz, CD3OD): 20.95
(CH3), 21.59 (C-3ꢀ), 34.25 (C-2ꢀ), 65.48 (C-6), 71.86 (C-4), 74.01
(C-5), 74.36 (C-3), 80.93 (C-2), 90.74 (C-1), 170.96 (
C ), 174.85
(C-1ꢀ). HRMS m/z [M+H]+ calc. 535.14201, measured 535.14162.
2.4. ˇ-N-Acetylhexosaminidase activity and inhibition assays
␤-N-Acetylhexosaminidase from the filamentous fungus
Talaromyces flavus CCF 2686 was obtained by heterologous expres-
sion in Pichia pastoris and purified by cation exchange chromatog-
raphy as described previously [34]. ␤-N-Acetylhexosaminidase
activity was assayed spectrophotometrically on microtiter
plates using 0.25 mM p-nitrophenyl 2-acetamido-2-deoxy-
␤-d-glucopyranoside (pNP-GlcNAc) as substrate, 50 mM
citrate-phosphate buffer pH 7 and 2 mU of the respective ␤-
N-acetylhexosaminidase, according to the published procedure

132
J. Krejzová et al. / Journal of Molecular Catalysis B: Enzymatic 87 (2013) 128–134
Scheme 2. Procedure A: C-6-monoacyl derivatives of NAG-thiazoline prepared by lipase catalysis.
to induce enhanced biological activity relative to the simple mono-
valent units [36]. In our case, we aimed at the improvement of the
␤-N-acetylhexosaminidase inhibition effect of the NAG-thiazoline
dimers caused by their divalency.
and variable concentration of the potential inhibitors were per-
formed at first. These measurements revealed atypical inhibition
pattern with stronger inhibition effect observed at lower con-
centration of the inhibitor, while in case of concentration of the
inhibitor reaching the concentration of the substrate (0.25 mM)
no inhibition was observed (data not shown). Moreover, at higher
substrate concentrations (0.5 mM and higher) no inhibition was
observed at any concentration of the tested compounds.
The divalent compounds 14 and 15 were prepared by linking
two units of NAG-thiazoline in the C-6 position via the diester
bridge (Scheme 3); the dicarboxylic acid with variable chain length
was used as the spacer. After the optimization of the solvent and
reaction conditions, the reaction of NAG-thiazoline and divinyl
dodecanedioate was performed in dry acetone at 45 ◦C, 250 rpm
in the presence of lipase Novozym 435 with the optimal molar
ratio of the divinyl dodecanedioate to NAG-thiazoline of 2.5. The
formation of the monoester of NAG-thiazoline was observed imme-
diately after the reaction was started, and after 1 h the target diester
of NAG-thiazoline appeared in the reaction mixture as well (TLC
EtOAc/MeOH 9/1). However, after two days of the reaction some
decomposition of the product was noticed, as NAG-thiazoline was
detected in the reaction mixture. Thus, the optimization of the
reaction time was crucial to prevent hydrolysis of the ester bond
formed. The isolated products were fully characterized and their
dimeric structures confirmed.
These results suggested that NAG-thiazoline and its deriva-
tives were not competitive inhibitors of this fungal enzyme as
we were expecting, thus, the more detailed Michaelis–Menten
analysis of enzyme kinetics was necessary to obtain more infor-
mation. Surprisingly, the results of these experiments were even
more puzzling, as no classical inhibition type could be assigned
to the obtained data. Tables 1 and 2 present the KM and Vmax
values (substrate pNP-GlcNAc), respectively, for the individual
compounds (1, 7, 9, 10, 14 and 15) at various concentrations;
obviously, the only trend that could be estimated was the gen-
eral increase of KM at higher inhibitor concentrations, while the
Vmax values irregularly oscillated. To avoid any errors all inhibitory
experiments were performed in five parallels. Our proposed con-
clusion is a kind of mixed inhibition originating in competitive
inhibition, as supported by the increasing KM values. To obtain
a more descriptive view on the obtained inhibition patterns, we
constructed the linearized Lineweaver–Burk plots, which revealed
that only compound 1 could be considered a competitive inhibitor
of the fungal ␤-N-acetylhexosaminidase. The plots of the other
compounds referred to non-competitive (7), un-competitive (10)
and also mixed (9, 14, 15) inhibition types. The more expressive
graphical visualization of these results together with the proposed
model of the observed phenomenon is presented in Supplementary
information online. However, this is the first report on the GH20 ␤-
N-acetylhexosaminidase, which did not display clearly competitive
3.2. Inhibition of ˇ-N-acetylhexosaminidase by the prepared
NAG-thiazoline derivatives
The inhibition effects of NAG-thiazoline 1 and its derivatives
(7–11, 14, 15) prepared by the lipase-catalyzed reactions were
tested using ␤-N-acetylhexosaminidase from the filamentous fun-
gus Talaromyces flavus CCF 2686. This enzyme is a representative of
eukaryotic ␤-N-acetylhexosaminidases with great synthetic abil-
ities; however, it had not been tested with inhibitors before.
Preliminary tests determining the relative enzyme activity at the
constant concentration of the enzyme and substrate (pNP-GlcNAc)
Scheme 3. Preparation of NAG-thiazoline diesters.

J. Krejzová et al. / Journal of Molecular Catalysis B: Enzymatic 87 (2013) 128–134
133
Table 1
KM values (mM) of ␤-N-acetylhexosaminidase-catalyzed reactions measured in the presence of the individual compounds in various concentrations using pNP-GlcNAc as
substrate.
CI (mM)
1
7
9
10
14
15
0
0.202 0.023
0.202 0.023
0.202 0.023
0.202 0.023
0.293 0.035
0.287 0.023
0.278 0.026
0.239 0.026
0.153 0.017
n.d.a
0.202 0.023
0.300 0.006
0.272 0.013
0.433 0.019
0.596 0.034
0.831 0.053
n.d.a
0.202 0.023
0.312 0.012
0.268 0.014
0.300 0.009
n.d.a
0.001
0.05
0.1
0.5
1
0.257 0.011
0.486 0.033
0.625 0.054
1.211 0.166
2.055 0.373
2.570 0.282
0.361 0.024
0.388 0.031
0.245 0.020
0.285 0.020
0.435 0.029
0.483 0.037
0.271 0.020
0.272 0.018
0.201 0.009
0.421 0.018
0.417 0.035
0.617 0.048
0.572 0.048
0.479 0.025
2
a
not determined.
Table 2
Vmax values (mmol/l/min) of ␤-N-acetylhexosaminidase-catalyzed reactions measured in the presence of the individual compounds in various concentrations using pNP-
GlcNAc as substrate.
CI (mM)
1
7
9
10
14
15
0
0.059 0.009
0.059 0.009
0.059 0.009
0.059 0.009
0.088 0.016
0.074 0.009
0.068 0.010
0.051 0.008
0.036 0.005
n.d.a
0.059 0.009
0.077 0.002
0.066 0.005
0.083 0.006
0.073 0.008
0.072 0.010
n.d.a
0.059 0.009
0.087 0.005
0.076 0.006
0.075 0.004
n.d.a
0.001
0.05
0.1
0.5
1
0.070 0.004
0.062 0.008
0.057 0.010
0.034 0.011
0.028 0.013
0.017 0.005
0.093 0.010
0.092 0.012
0.064 0.008
0.051 0.005
0.051 0.006
0.065 0.010
0.084 0.009
0.075 0.007
0.053 0.003
0.082 0.006
0.069 0.010
0.080 0.012
0.077 0.012
0.063 0.006
2
a
not determined.
inhibition with NAG-thiazoline, the known typical competitive
inhibitor of this group of enzymes [18,22] based on analogy with
the reaction transition state NAG-oxazoline. In general the inhi-
bition of the fungal ␤-N-acetylhexosaminidases has been studied
only scarcely so far. The observed behaviour could be explained
by a model assuming binding of two molecules of a compound to
the enzyme; one inhibiting and one activating (see Supplementary
material); the relevance of the proposed model has been clarified
at least with compounds 9 and 14.
References
[1] G. Carrea, S. Riva, Angew. Chem. Int. Ed. 39 (2000) 2226–2254.
[2] S. Riva, in: G. Carrea, S. Riva (Eds.), Organic Synthesis with Enzymes in Non-
Aqueous Media, Wiley-VCH Verlag GmbH, Weinheim, Germany, 2008, pp.
145–167.
[3] A.M. Gumel, M.S.M. Annuar, T. Heidelberg, Y. Christi, Process Biochem. 46
(2011) 2079–2090.
[4] Y. Shi, J. Li, Y. Chu, J. Chem. Technol. Biotechnol. 86 (2011) 1457–1468.
[5] R. Ye, S.H. Pyo, D.G. Hayes, J. Am. Oil Chem. Soc. 87 (2010) 281–293.
[6] Q. Wu, A. Xia, X. Lin, J. Mol. Catal. B: Enzym. 54 (2008) 76–82.
[7] P. Magrone, F. Cavallo, W. Panzeri, D. Pasarella, S. Riva, Org. Biomol. Chem. 8
(2010) 5583–5590.
[8] K. Purchartová, P. Marhol, R. Gazˇák, D. Monti, S. Riva, M. Kuzma, V. Krˇen, J. Mol.
4. Conclusion
Catal. B: Enzym. 71 (2011) 119–123.
[9] P. Bojarová, K. Krˇenek, M. Kuzma, L. Petrásková, K. Bezousˇka, D.J. Namdjou, L.
Elling, V. Krˇen, J. Mol. Catal. B: Enzym. 50 (2008) 69–73.
[10] K. Slámová, R. Gazˇák, P. Bojarová, N. Kulik, R. Ettrich, H. Pelantová, P. Sedmera,
V. Krˇen, Glycobiology 20 (2010) 1002–1009.
[11] P. Bojarová, K. Slámová, K. Krˇenek, R. Gazˇák, N. Kulik, R. Ettrich, H. Pelantová, M.
Kuzma, S. Riva, D. Adámek, K. Bezousˇka, V. Krˇen, Adv. Synth. Catal. 353 (2011)
2409–2420.
A set of monomeric and dimeric C-6-acylated analogues of
NAG-thiazoline was prepared in lipase-catalyzed reactions; the
dimeric thiazoline esters were reported for the first time. These
compounds were tested as potential inhibitors of a fungal ␤-N-
acetylhexosaminidase resulting in all kinds of known inhibition
patterns, even though the parent compound NAG-thiazoline is a
well-known competitive inhibitor of this enzyme class. The inhi-
bition capabilities of the prepared modified NAG-thiazolines will
be further studied with other GH20 and GH84 glycosidases from
various sources.
[12] D. Kavan, M. Kubícˇková, J. Bíly´, O. Vaneˇk, K. Hofbauerová, H. Mrázek, D.
ˇ
Rozbesky´, P. Bojarová, V. Krˇen, L. Zídek, V. Sklenárˇ, K. Bezousˇka, Biochemistry
49 (2010) 4060–4067.
[13] R.G. Korneluk, D.J. Mahuran, K. Neote, M.H. Klavins, B.F. O’Dowd, M. Tropak,
H.F. Willard, M.J. Anderson, J.A. Lowden, R.A. Gravel, J. Biol. Chem. 261 (1986)
8407–8413.
[14] R.L. Proia, Proc. Natl. Acad. Sci. U.S.A. 85 (1988) 1883–1887.
[15] S.T. Hepbildikler, R. Sandhoff, M. Kolzer, R.L. Proia, K. Sandhoff, J. Biol. Chem.
277 (2002) 2562–2572.
[16] C.R. Torres, G.W. Hart, J. Biol. Chem. 259 (1984) 3308–3317.
[17] Y. Gao, L. Wells, F.I. Comer, G.J. Parker, G.W. Hart, J. Biol. Chem. 276 (2001)
9838–9845.
[18] N. Comtesse, E. Maldener, E. Meese, Biochem. Biophys. Res. Commun. 283
(2001) 634–640.
[19] D.J. Mahuran, Biochim. Biophys. Acta 1455 (1999) 105–138.
[20] W.B. Dias, G.W. Hart, Mol. Biosyst. 3 (2007) 766–772.
[21] J.A. Hanover, M.W. Krause, D.C. Love, Biochim. Biophys. Acta 1800 (2009) 80–95.
[22] M.S. Macauley, G.E. Whitworth, A.W. Debowski, D. Chin, D.J. Vocadlo, J. Biol.
Chem. 280 (2005) 25313–25322.
Acknowledgements
This work was supported by the grant P207/11/0629 from the
Czech Science Foundation and by project RVO61388971 (Institute
of Microbiology research concept). We would like to thank Dr. Josef
Cvacˇka from the Institute of Organic Chemistry and Biochemistry
AS CR for the measurement of the HRMS spectra of the vinyl esters.
ˇ
ESF COST Chemistry Actions CM0701 “CASCAT” (MSMT OC09045)
and CM1002 “MultiGlycoNano” are acknowledged.
[23] C. Martinez-Fleites, J.E. Korczynska, G.J. Davies, M.J. Cope, J.P. Turkenburg, E.J.
Taylor, Carbohydr. Res. 344 (2009) 1753–1757.
[24] S.A. Yuzwa, M.S. Macauley, J.E. Heinonen, X. Shan, R.J. Dennis, Y. He, G.E. Whit-
worth, K.A. Stubbs, E.J. McEachern, G.J. Davies, D.J. Vocadlo, Nature Chem. Biol.
4 (2008) 483–490.
Appendix A. Supplementary data
[25] M.S. Macauley, D.J. Vocadlo, Biochim. Biophys. Acta – Gen. Subjects 1800 (2010)
107–121.
[26] S. Knapp, D. Vocadlo, Z. Gao, B. Kirk, J. Lou, S.G. Withers, J. Am. Chem. Soc. 118
(1996) 6804–6805.
Supplementary data associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.molcatb.
2012.10.016.

134
J. Krejzová et al. / Journal of Molecular Catalysis B: Enzymatic 87 (2013) 128–134
[27] B.L. Mark, D.J. Mahuran, M.M. Cherney, D. Zhao, S. Knapp, M.N.G. James, J. Mol.
Biol. 327 (2003) 1093–1109.
[31] N. Armesto, M. Ferrero, S. Fernández, V. Gotor, J. Org. Chem. 68 (2008)
5784–5787.
[28] S. Knapp, M. Abdo, K. Ajayi, R.A. Huhn, T.J. Emge, E.J. Kim, J.A. Hanover, Org. Lett.
9 (2007) 2321–2324.
[32] G.C.R. Ellis-Davies, A. Gilbert, P. Heath, J.C. Lane, J.V. Warrington, D.L. Westover,
J. Chem. Soc. Perkin Trans. 2 (1984) 1833–1841.
[29] B. Amorelli, C. Yang, B. Rempel, S.G. Withers, S. Knapp, Bioorg. Med. Chem. Lett.
18 (2008) 2944–2947.
[30] P. Compain, C. Decroocq, J. Iehl, M. Holler, D. Hazelard, T.M. Barra-
gan, C. Ortiz Mellet, J.-F. Nierengarten, Angew. Chem. Int. Ed. 49 (2010)
5753–5756.
[33] R. Sadychsade, Chem. Abstr. 63 (1965) 14984.
[34] K. Slámová, P. Bojarová, D. Gerstorferová, B. Fliedrová, J. Hofmeisterová, M.
Fiala, P. Pompach, V. Krˇen, Protein Expr. Purif. 82 (2012) 212–217.
[35] P. Rashidi-Ranjbar, F. Piri, Molecules 4 (1999) 135–138.
[36] G. Bérubé, Curr. Med. Chem. 13 (2006) 131–154.