Glycosynthases from Thermotoga neapolitana β-glucosidase 1A: A comparison of α-glucosyl fluoride and in situ-generated α-glycosyl formate donors

Article abstract of DOI:10.1016/j.molcatb.2014.05.021

TnBgl1A from the thermophile Thermotoga neapolitana is a dimeric β-glucosidase that belongs to glycoside hydrolase family 1 (GH1), with hydrolytic activity through the retaining mechanism, and a broad substrate specificity acting on β-1,4-, β-1,3- and β-1,6-linkages over a range of glyco-oligosaccharides. Three variants of the enzyme (TnBgl1A-E349G, TnBgl1A-E349A and TnBgl1A-E349S), mutated at the catalytic nucleophile, were constructed to evaluate their glycosynthase activity towards oligosaccharide synthesis. Two approaches were used for the synthesis reactions, both of which utilized 4-nitrophenyl β-d-glucopyranoside (4NPGlc) as an acceptor molecule: the first using an α-glucosyl fluoride donor at low temperature (35 °C) in a classical glycosynthase reaction, and the second by in situ generation of the glycosyl donor with (4NPGlc), where formate served as the exogenous nucleophile under higher temperature (70 °C). Using the first approach, TnBgl1A-E349G and TnBgl1A-E349A synthesized disaccharides with β-1,3-linkages in good yields (up to 61%) after long incubations (15 h). However, the GH1 glycosynthase Bgl3-E383A from a mesophilic Streptomyces sp., used as reference enzyme, generated a higher yield at the same temperature with both a shorter reaction time and a lower enzyme concentration. The second approach yielded disaccharides for all three mutants with predominantly β-1,3-linkages (up to 45%) but also β-1,4-linkages (up to 12.5%), after 7 h reaction time. The TnBgl1A glycosynthases were also used for glycosylation of flavonoids, using the two described approaches. Quercetin-3-glycoside was tested as an acceptor molecule and the resultant product was quercetin-3,4′-diglycosides in significantly lower yields, indicating that TnBgl1A preferentially selects 4NPGlc as the acceptor.

Full text of DOI:10.1016/j.molcatb.2014.05.021

Accepted Manuscript
Title: Glycosynthases from Thermotoga neapolitana
␤-glucosidase 1A: A comparison of ␣-glucosyl fluoride and in
situ-generated ␣-glycosyl formate donors
´
Author: Tania Pozzo Merichel Plaza Javier Romero-Garcıa
Magda Faijes Eva Nordberg Karlsson Antoni Planas
PII:
S1381-1177(14)00171-4
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.molcatb.2014.05.021
MOLCAB 2969
To appear in:
Journal of Molecular Catalysis B: Enzymatic
Received date:
Revised date:
Accepted date:
28-1-2014
26-5-2014
27-5-2014
´
Please cite this article as: T. Pozzo, M. Plaza, J. Romero-Garcia, M. Faijes,
E.N. Karlsson, A. Planas, Glycosynthases from Thermotoga neapolitana rmbeta-
glucosidase 1A: A comparison of rmalpha-glucosyl fluoride and in situ-generated
rmalpha-glycosylformatedonors, JournalofMolecularCatalysisB:Enzymatic(2014),
http://dx.doi.org/10.1016/j.molcatb.2014.05.021
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

1
2
3
Glycosynthases from Thermotoga neapolitana β-glucosidase 1A: A
comparison of α-glucosyl fluoride and in situ-generated α-glycosyl
formate donors
4
5
6
Tania Pozzoa, Merichel Plazab, Javier Romero-Garcíac, Magda Faijesc, Eva Nordberg
Karlssona*, Antoni Planasc
7
8
aBiotechnology, Lund University, P.O. Box 124, SE-22100 Lund, Sweden
9
10
bDepartment of Chemistry, Centre for Analysis and Synthesis, P.O. Box 124, SE-22100,
Lund, Sweden
11
12
cLaboratory of Biochemistry, Institut Quimíc de Sarrià, Universitat Ramon Llull, Via Augusta
390, E-08017 Barcelona, Spain
13
14
*Corresponding Author. Tel.: +46-46-2224626 E-mail address: eva.nordberg-
karlsson@biotek.lu.se
15
1
Page 1 of 32

15
16
Highlights
17
18
19
•
•
•
Selective synthesis with α-glucosyl fluoride donors and thermophilic
glucosynthase.
20
21
22
Synthesis with in situ generated glycosyl-donor also confirmed, but less
selective.
23
24
25
Flavonoids are alternative acceptors for the synthase with both types of
donors.
26
27
28
29
Abstract
30
31
TnBgl1A from the thermophile Thermotoga neapolitana is a dimeric β-glucosidase
that belongs to glycoside hydrolase family 1 (GH1), with hydrolytic activity through
the retaining mechanism, and a broad substrate specificity acting on β-1,4-, β-1,3- and
β-1,6-linkages over a range of glyco-oligosaccharides. Three variants of the enzyme
(TnBgl1A_E349G, TnBgl1A_E349A and TnBgl1A_E349S), mutated at the catalytic
nucleophile, were constructed to evaluate their glycosynthase activity towards
oligosaccharide synthesis. Two approaches were used for the synthesis reactions, both
of which utilized 4-nitrophenyl β-D-glucopyranoside (4NPGlc) as an acceptor
molecule: the first using an α-glucosyl fluoride donor at low temperature (35°C) in a
classical glycosynthase reaction, and the second by in situ generation of the glycosyl
donor with (4NPGlc), where formate served as the exogenous nucleophile under
higher temperature (70° C). Using the first approach, TnBgl1A_E349G and
TnBgl1A_E349A synthesised disaccharides with β-1,3-linkages in good yields (up to
61%) after long incubations (15h). However, the GH1 glycosynthase Bgl3_E383A
from a mesophilic Streptomyces sp., used as reference enzyme, generated a higher
yield at the same temperature with both a shorter reaction time and a lower enzyme
concentration. The second approach yielded disaccharides for all three mutants with
predominantly β-1,3-linkages (up to 45%) but also β-1,4-linkages (up to 12.5%), after
7h reaction time. The TnBgl1A glycosynthases were also used for glycosylation of
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
2
Page 2 of 32

49
50
51
flavonoids, using the two described approaches. Quercetin-3-glycoside was tested as
an acceptor molecule and the resultant product was quercetin-3,4´-diglycosides in
significantly lower yields, indicating that TnBgl1A preferentially selects 4NPGlc as
the acceptor.
52
53
54
55
Keywords: β-glucosidase; α-glucosyl fluoride; Formate; 4-nitrophenyl β-D-
glucopyranoside; Quercetin
56
3
Page 3 of 32

56
57
1. Introduction
58
59
60
61
Glycoside hydrolases (GH) are enzymes widespread in nature, with the main function
of breaking down glycosidic bonds within a diverse array of biological processes
including cellular metabolism, defence mechanisms and the degradation of biomass
[1]. In order to better understand GHs, they have been classified into more than 132
families based on amino-acid sequence homology data that has been archived within
the Carbohydrate Active Enzyme database [2]. The β-glucosidase TnBgl1A from
Thermotoga neapolitana belongs to glycoside hydrolase family 1 (GH1 and displays
both the typical TIM (β/α)8 barrel-fold characteristic of the family and hydrolytic
activity over a broad range of glyco-oligosaccharides by the retaining mechanism [3].
62
63
64
65
66
67
68
69
70
71
The main hydrolytic function of GHs in nature can be changed in vitro towards glyco-
oligosaccharide synthesis, and the formation of glycosidic linkages can occur via
reverse hydrolysis (thermodynamically controlled) or transglycosylation (kinetically
controlled) [4,5]. However, the major obstacle and disadvantage to using GHs for
glyco-oligosaccharide synthesis lies in the fact that hydrolytic activity remains present
as the main function of these enzymes. In general, the use of GHs for these two
synthetic reactions results in low yields due to the frequent hydrolysis of glycosidic
linkages formed during oligosaccharide synthesis [6,7,8].
72
73
74
75
76
77
78
79
80
81
In the investigation for efficient methodologies on enzymatic synthesis of glycosidic
linkages in vitro, an alternative approach has emerged with the creation of
hydrolytically-inactive-nucleophile, mutated variants called “glycosynthases”, which
efficiently catalyse the glycoside linkage formation as a primary function and are
capable of achieving high yields for oligosaccharide synthesis [9]. Glycosynthases are
hence engineered glycoside hydrolases in which the catalytic nucleophile (Glu or Asp
residues) has been replaced by a non-nucleophilic residue with a shorter side chain,
commonly Gly, Ala, or Ser [10]. They are inactive hydrolases but efficiently catalyze
glycosyl transfer to an acceptor when using an activated glycosyl fluoride donor
(GlcF) with the opposite anomeric configuration than the original substrate of the
parental wild type hydrolase reaction. [11,12]. An alternative glycosynthase reaction
for thermostable glycosynthases was developed using external nucleophiles, such as
formate, which mimics the glycosyl-enzyme intermediate mechanism that utilizes
activated substrates [13].
82
83
84
85
86
87
88
4
Page 4 of 32

89
90
91
In this study, thermostable glycosynthases from Thermotoga neapolitana (TnBgl1A)
were created, by mutating the catalytic nucleophile E349 to either G, A or S, and
characterised for use in gluco-oligosaccharide synthesis while following two different
approaches: a) using α-glycosyl fluoride as a donor and b) using sodium formate as an
exogenous nucleophile for in situ generation of an α-glycosyl-formate donor. In the
former reaction, Bgl3_E383A from Streptomyces sp., reported as an efficient
glycosynthase with regioselectivity for β-1,3-linkage [9], was also used to provide a
mesophilic enzyme for comparison. In a follow up study, flavonoids were studied as
alternative acceptor molecules for enzymatic glycosylation by thermostable
glycosynthases under high temperature.
92
93
94
95
96
97
98
99
100
101
2. Materials and Methods
2.1 Chemicals
102
103
104
105
106
Commercial substrates, 4-Nitrophenyl β-D-glucopyranoside (4NPGlc), quercetin-3-β-
D-glucoside, quercetin, and sodium formate were supplied by Sigma-Aldrich
(Steinheim, Germany). Quercetin-3,4’-di-O-β-D-glucopyranoside was purchased from
Polyphenols Laboratories AB (Sandnes, Norway). Formic acid was provided from
Merck (Darmstadt, Germany) and methanol by Scharlau (Barcelona, Spain).
107
108
109
110
The α-glycosyl fluoride, donor molecule was chemically synthesized as described
previously [9]. Briefly, peracetylated glucose was fluorinated with hydrogen fluoride
pyridine, followed by de-O-acetylation with sodium methoxide in methanol, which
resulted in high yields of α-glycosyl fluoride (more than 80%).
111
112
2.2 Engineering TnBgl1A glycosynthases
113
114
115
116
117
118
119
120
Three inactive-nucleophile variants, TnBgl1A_E349G, TnBgl1A_E349A and
TnBgl1A_E349S, were constructed via site directed mutagenesis using both the
plasmid encoding the β-glucosidase TnBgl1A gene from T. neapolitana as the
template, and the mutagenic primers (Eurofins MWG Operon, Ebersberg, Germany)
described in Table 1. PCR reactions were run in a T-Gradient thermocycler
(Biometra, Göttingen Germany) using iProof high fidelity DNA polymerase
(iProofTM High-Fidelity master mix Bio-Rad laboratories, California, USA); the
thermal cycling protocol was as follows: an initial denaturing step of 30 s at 95° C,
5
Page 5 of 32

121
followed by 30 cycles of denaturing 30 s at 95° C, annealing 60 s at 55° C and
extension of 5 min 30 s at 68° C with a final extension of 10 min at 68° C.
122
123
124
125
126
127
128
129
130
131
PCR reactions were digested with DpnI (New England, Biolabs) for one hour to
remove the template and transformed into electrocompetent Escherichia coli
NovaBlue (Novagen) using the GenePulser II (Bio-Rad), immediately incubated for 1
hour at 37° C and spread on LB agar containing 100 μg/mL ampicillin. Successfully
transformed E.coli colonies were selected, and an overnight inoculum was prepared
for plasmid extraction in 5 ml LB media with ampicillin (100 μg/mL). Plasmids were
purified using the Zyppy plasmid miniprep kit (Zymo Research, CA, USA), and sent
for sequencing using vector specific primers T7 and pET-RP supplied by the
sequencing facility at GATC Biotech (Konstanz, Germany) to obtain complete
coverage of the entire β-glucosidase genes. The mesophilic glucosynthase,
Bgl3_E383A, from Streptomyces sp. was constructed previously by Vallmitjana et al.
[14], where the site directed mutagenesis was carried out by using a mutagenic /
reverse universal primer pair,and an overlapping PCR method that was described
previously by Juncosa et al. [15].
132
133
134
135
136
137
138
2.3 Protein production and purification
139
140
141
Inactive nucleophile variants and wild type TnBgl1A were transformed into
competent E.coli BL21 (DE3) (Novagen) for production. Production was carried out
in a bioreactor of 2.5 L at 37° C, pH 7, using a defined medium, described in
Ramchuran [16] with 100 μg/mL ampicillin. Enzyme production was induced with
0.1mM of IPTG when the OD at 620nm reached 0.8, and cultivation continued up to 3
hours. Cultures were centrifuged (10000 × g, 10 min, 4° C), and cell pellets were
dissolved in binding buffer (20 mM imidazole, 20 mM Tris-HCl, 0.75 M NaCl,
pH 7.5) and lysed by sonication for 3×2 min using a 14 mm titanium probe sound
intensity of 60% and a cycle of 0.5 (UP400S, Dr. Hielscher) followed by a final
centrifugation of 30 min, 39000 × g, 4° C as described by Turner [17].
142
143
144
145
146
147
148
149
150
151
The supernatant was heat treated at 70° C for 30 min followed by centrifugation
(10000 × g, 20 min) and the resulting supernatant containing the His-tagged enzyme
was purified by immobilised metal ion affinity chromatography (IMAC). The IMAC
was performed using a HiTrap affinity 5 ml column (GE health care, Germany) with a
152
6
Page 6 of 32

153
154
155
156
157
copper ligand on an ÄKTA prime system (Amersham Biosciences, Uppsala, Sweden).
Fractions containing the purified enzyme were pooled and dialyzed against 20 mM
citrate-phosphate buffer, pH 5.6, overnight using a dialysis membrane with a cut-off
of 3,500 Da (Spectrum laboratories, Rancho Dominguez, CA, USA). The dialysed
proteins were then stored at 4° C for future use.
158
159
160
161
The mesophilic Bgl3_E383A was expressed constitutively without the addition of
IPTG in 200 ml of LB medium with 100 μg/ml. Mesophilic Bgl3_E383A was then
purified by IMAC on HiTrap Chelating Sepharose (Pharmacia) using nickel as
described by Vallmitjana [14], followed by dialysis of pure Bgl3_E383A with 20 mM
phosphate buffer at pH 7. Enzyme purity was confirmed by visualisation with SDS-
PAGE and the protein concentration was estimated by measuring absorbance at 280
nm with the NanoDrop 1000 (Thermo Fisher Scientific).
162
163
164
165
166
2.4 Hydrolytic activity
167
168
169
170
171
Initial velocities were determined to estimate hydrolytic activity of TnBgl1A and
residual activity for inactive nucleophilic variants at two different pHs (5.6 and 7) and
temperatures (35° C and 80° C). The reaction began by adding 20 μl of the enzyme
(from 0.01 to 0.5 μM) to a pre-incubated 2 mM 4NPGlc in 50 mM citrate-phosphate
buffer at both pH 5.6 and 7 under both thermal conditions. Extinction coefficient
values were: εM 35°C pH5.6 1898 M-1cm−1; εM 80°C pH5.6 3789 M-1cm−1; εM 35°C pH7 14905
M-1cm−1; εM 80°C pH7 15203 M-1cm−1.
172
173
174
175
2.5 Chemical reactivation using Sodium Azide
176
177
178
179
180
181
Chemical reactivation experiments of inactive nucleophilic variants using sodium
azide (from 0.5 to 3 M) as external nucleophiles were first evaluated on the 96 well
plate assay at 35° C for 30 min (Thermo Scientific Multiskan GO UV/Vis
Microplate). Different 4NPGlc concentrations were used (0.2 to 4 mM) in a 50 mM
citrate-phosphate buffer at pH 5.6 alongside a fixed enzyme concentration of 0.08 μM
within a total volume of 250 μl. Blanks for each 4NPGlc concentration were run to
assure that spontaneous substrate hydrolysis was negligable.
182
183
7
Page 7 of 32

184
185
186
2.6 Glycosynthase activity using α-glycosyl fluoride donor
187
188
189
190
191
Glycosynthase activity reactions were run, using 1 mM α-glucosyl fluoride (GlcF) and
5 mM 4NPGlc in a 50 mM buffer at pH 7. The reaction was pre-incubated for 5 min
before adding the enzyme (between 0.5 to 2.5 μM) to achieve a final volume of
300 μL for the total reaction. The reactions were monitored upon addition of the
enzyme; aliquots of 10 μL were withdrawn every hour for 16 hours in the case of
thermophilic glycosynthases, and every 15 min over a 1-hour time period for the
mesophilic glycosynthase [9]. Samples were diluted with deionized water (1:10)
before injection in the High Performance Liquid Chromatograph (HPLC) for
monitoring oligosaccharide synthesis.
192
193
194
195
196
197
2.7 Glycosynthase activity by rescue with exogenous nucleophile
198
199
200
201
Glycosynthase activity reactions were run for the thermophilic glycosynthases using
6 mM 4NPGlc in 50 mM citrate-phosphate over a pH-range of 3 to 7 with different
concentrations of the exogenous nucleophile sodium formate (0.1 to 2 M). The
reaction was pre-incubated for 5 min at 70° C before adding the enzyme (0.5 to
2.7 μM). Afterwards, monitoring began by withdrawing aliquots of 20 μL every hour
for 7 hours. The blanks that were run are as follows: blank 1, reaction without enzyme
to check spontaneous hydrolysis of the substrate with an external nucleophile and
blank 2, reaction without external nucleophile to check for the possibility of
spontaneous transglycosylation reactions. All aliquots were diluted with deionized
water (1:10 and 1:100) before injection to HPLC and High-Performance Anion-
Exchange Chromatography with Pulsed Amperometric Detection (HPAEC-PAD).
202
203
204
205
206
207
208
209
210
2.8 Monitoring enzymatic synthesis of oligosaccharides.
211
212
213
214
HPLC was used for monitoring oligosaccharide formation using a C18 column
(Novapak, 4 um Waters); identification of the linkage type was achieved by co-
injection of standards followed by UV detection at 300 nm. The mobile phase used
was 14% methanol (MeOH) in water at 1 mL/min flow rate. Reactions incubated after
8
Page 8 of 32

215
216
217
218
16 h were visualised using thin layer chromatography (TLC). Sample was loaded onto
a silica gel TLC (Silica 60 F254, Merck, Darmstadt, Germany) and separated using
ethyl acetate/MeOH/water (7:2.5:1) as mobile phase. Plates were developed with
sulfuric acid/MeOH/water (10:45:45) followed by heating at 120° C [9].
219
220
221
HPAEC-PAD was used to measure glucose liberation during the glycosynthase
reaction, which was quantified by integration of glucose peak areas based on a
standard curve (0.5 to 10 μM glucose), using a Dionex ICS-5000, Ion Chromatograph
equipment (Dionex Corporation, Sunnyvale, CA, USA). The column was CarboPac
PA-200 anion-exchange column (3 x 150 mm and 3 x 30 mm CarboPac PA-200 guard
column). Glucose was eluted using a flow rate of 0.5 ml min-1 by a gradient of
eluents: Constant 100 mM NaOH 0 to 20 min, Gradient 0-15 min of 0-34mM NaAc
from 15 to 20 min, then constant 100 mM NaAc. Chromeleon 7 software was used to
control the system and analyse the data.
222
223
224
225
226
227
228
229
2.9 Enzymatic glycosylation of flavonoids
230
231
Enzymatic glycosylations of the flavonoid molecules (quercetin and quercetin-3-β-D-
glucoside) were carried out at donor:acceptor-ratios of 1:3 using either 4NPGlc or
2NPGlc as a donor and 2 M of sodium formate as exogenous nucleophiles, at 1 μM
enzyme concentration in 50 mM citrate phosphate buffer at pH 5.6. Reactions were
pre-incubated for 5 min at 70° C before adding the enzyme, after which monitoring
began by withdrawing aliquots of 10 μL every hour during 16 hrs. The sample was
diluted with MeOH (1:10) before HPLC injection. The blanks measured were as
follows: blank 1, reaction without enzyme to check for spontaneous hydrolysis of the
substrate with an external nucleophile and blank 2, reaction without an external
nucleophile to check for the possibility of spontaneous transglycosylation reactions.
The classical glycosynthase reaction using both GlcF and Q3Glc as the acceptor
molecules was also tested in a 1:3 ratio at 35°C.
232
233
234
235
236
237
238
239
240
241
242
243
2.10 Monitoring enzymatic glycosylation of flavonoids
244
245
246
Glycosylated forms were detected using an UltiMate-3000® HPLC system from
Dionex (Thermo Fisher, Germering, Germany) that consisted of an online degasser, a
quaternary solvent pump, an autosampler with cooler, column oven, and photo diode
9
Page 9 of 32

247
248
249
250
251
array detector (DAD) with scanning capabilities, all controlled by Chromeleon 6.80
(Thermo Fisher) software. The detection wavelengths were set at 280, 300 and 350
nm. A porous-shell fused core Ascentis Express C18 (150 mm x 2.1 mm, 2.7 μm.)
from Supelco (Bellefonte, PA, USA) was used as an analytical column for LC
separation. The column temperature was 50 °C and injection volume was 2 μL, the
vial tray was held at 4 °C. Mobile phases consisted of (A) water with 0.5% (v/v)
formic acid (B) methanol with 0.5% (v/v) formic acid in a gradient elution analysis
programmed as follows: 0 min, 5% (B); 0-5 min, 5% (B); 5-35 min, 40% (B); 35-40
min, 40% (B); at a flow rate of 300 μL/min.
252
253
254
255
256
257
258
259
260
261
The glycosylated flavonoids were further identified by MS/MS, and the equipment
used was a liquid chromatograph Accela equipped with DAD coupled to a Thermo
Exactive orbitrap mass spectrometer (MS) via an electrospray interface (ESI)
(Thermo Fisher, Germering, Germany). The MS was operated under ESI negative
mode. Tuning and optimization were carried out using a direct injection of quercetin
3-glucoside standard (0.004 mmol) at 0.750 mL/min. The carrier gas was ultrahigh-
purity nitrogen.
262
263
264
265
266
267
268
269
270
271
2.11 Structural studies of β-glycosynthases
Docking experiments were prepared using the crystal structure of TnBgl1A_E349G
(PDB-code to be added). This structure was protonated at pH 5.6 with the server H++
[20]. An α-glucosyl fluoride (donor) molecule drawn using the server PRODRG [18],
was placed at the -1 subsite of theTnBgl1A_E349G structure by means of a
superimposition with the ligand (2-deoxy-2-fluoro-alpha-D-glucopyranose) of β-
glucosidase TmBglA from Thermotoga maritima (1OIN) and was subsequently
minimized using GROMACS [19]. The acceptor molecule 4NPGlc was obtained from
the crystal structure 3AI0, the quercetin 3-glucoside was drawn with PRODRG, and
both were used as ligands for a docking experiment where they were placed along
both the +1 and +2 subsites of the TnBgl1A_E349G structure. The docking was
realized by AutoDock VINA [20] and the chosen docked ligand was minimized with
both the protein and donor using GROMACS. The results and figures were visualized
using VMD [22].
272
273
274
275
276
277
278
10
Page 10 of 32

278
279
280
281
3. Results and Discussion
3.1 Engineering, expression and purification
The gene encoding the β-glucosidase, TnBgl1A from T. neapolitana, was previously
cloned into the pET-22b(+) vector under control of the T7/lac promoter for production
of the protein product in E.coli [23]. This construct was amplified in the methylating
E.coli strain Nova Blue, and was used as a template for engineering three inactive
nucleophile variants, TnBgl1A_E349G, TnBgl1A_E349A and TnBgl1A_E349S by
site directed mutagenesis. The mutations were inserted using a ligation independent
amplification of the template plasmid with overlapping mutagenic primers (Table 1),
followed by selective digestion of the methylated template. Later on, for production,
the separate plasmids encoding either one of the nucleophile variants or TnBgl1A
wild type (wt) were transformed into E.coli BL21 (DE3) and each line produced the
target protein in 2.5 l batch cultivations at 37° C, pH 7. Although overall expression
was high (around 50% of total protein), the major fraction of recombinant enzyme
was insoluble, resulting in a yield for the desired soluble target enzyme that accounted
for approximately 5% of the total soluble protein present in the extract after cell
disruption. There was no visible difference in the production yields between the wt
and the three nucleophile-mutated variants.
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
The high thermostability of the enzymes motivated a heat treatment step, which
resulted in approximately 50% purity, where more than 90% purity (as judged by
SDS-PAGE) was reached after IMAC purification using a copper ligand that bound
selectively to the C-terminal His6-tag that was fused to all of the enzymes. After the
two purification steps, approximately 50 mg of purified TnBgl1A was obtained per
litre of batch-culture.
302
303
304
3.2 Hydrolytic activity
305
306
307
308
309
The classical glycosynthase reaction, using fluorinated sugars, was performed at
temperatures no higher than 40º C and pH 7, due to lability of the fluorinated donor
[24,25]. Optimal conditions for TnBgl1A’s hydrolytic activity were previously
determined to be 80° C and pH 5.6 in Khan’s work [26]. The results displayed in
Figure 1 show that hydrolytic activity was evaluated under different conditions of
11
Page 11 of 32

310
311
temperature (80°C and 35°C) and pH (5.6 and 7), where the initial velocity of
TnBgl1A at 80°C, pH5.6 was considered to be 100%. When the temperature was
reduced to 35°C (keeping the optimal pH of 5.6) the activity decreased to 25%. The
effect of the pH was also significant within the interval, due to the fact that at 80°C an
increase in pH from 5.6 to 7 resulted in a 30% reduction in activity. This observed
level of enzymatic activity fell within the same range as the activity drop observed
upon the temperature reduction from 80°C to 35°C at a constant pH. Despite the only
10% enzymatic activity that remained under the conditions of 35ºC, pH 7, these
appeared to be appropriate for the glycosynthase reactions with mutated nucleophile
variants. This low activity level also explains the demand of higher enzyme
concentrations and longer incubation times (in addition to the fact that most
glycosynthases have lower specific activity compared to their primary wt-hydrolase
activity).
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
Any residual hydrolytic activities in the mutated nucleophile variants was also
assayed under the previously described experimental conditions, and resulted in either
little or no detectable activity, the results of which are presented in Table 2.
3.3 Chemical reactivation of mutated nucleophile variants with Sodium Azide
Chemical reactivation with sodium azide (NaN3) was performed next in order to
evaluate the possibility to utilize the mutated nucleophile variants as glycosynthases.
Different NaN3 concentrations (0.5 to 4 M) were assayed in an effort to restore
activity to the inactive nucleophile variants at 35° C by using 4NPGlc (from 0.2 to
3 mM) as a substrate in yielding the α-glucosyl azide product [9]. The results
indicated that 1 M NaN3 was the appropriate concentration for restoring maximum
activity to both TnBgl1A_E349A and TnBgl1A_E349S. The observed level of
reinstated activity by TnBgl1A_E349S was shown to be twice as high as for
TnBgl1A_E349A, whereas TnBgl1A_E349G failed to demonstrate any activity as
shown in Figure 2.
332
333
334
335
336
337
338
339
340
341
342
343
3.4 Glycosynthase activity using α-glucosyl fluoride donor
As all three mutated nucleophile variants TnBgl1A_E349A, TnBgl1A_E349S, and
TnBgl1A_E349G showed low residual hydrolytic activity (Table 2) and two
(TnBgl1A_E349A, TnBgl1A_E349S) of the three were reactivated with an exogenous
azide (Fig. 2), it was considered amenable to test all three as glycosynthases. The
12
Page 12 of 32

344
345
346
347
348
349
350
351
classical glycosynthase reaction mechanism, described in Fig.3, was carried out using
α-glucosyl fluoride and 4NPGlc as the donor and acceptor molecules respectively in a
ratio of 1:5.
All three glycosynthase variants (including TnBgl1A_E349G, which was not
chemically rescued with azide) synthesized a single disaccharide product that was
identified by HPLC and TLC as 4-nitrophenyl-β- laminaribioside (4NPLam), which is
an indication of a glycosynthase activity with selectivity towards a disaccharide with a
β-1,3 glycosidic linkage. In a parallel experiment, the well-known glycosynthase
Bgl3_E383A, derived from Streptomyces sp. [9] was assayed under similar
experimental conditions, producing substantial yields of 4NPLam (Table 3).
352
353
354
355
356
357
358
359
360
361
In a first trial (60 min duration at 35° C, pH 7), the nucleophile variants of TnBgl1A
seemed to have low initial rates and overall efficiencies compared to Bgl3_E383A,
which reached 100 % yield in 4NPLam production. However, this was not surprising
due to the fact that these thermostable enzymes were operating at least 45° C below
their activity optimum, where only 10% of the optimum activity was expected to be
achieved (Fig.1, Table 2).
In an attempt to improve the oligosaccharide yields first obtained at 0.5 µM enzyme,
the reaction was run at a higher enzyme concentration for a longer incubation time
(Table 3). After 15 hours, TnBgl1A_E349G (2.5 μM) improved to show yields up to
62%, TnBgl1A_E349A (1.7 μM) yielded 21% oligosaccharide, and TnBgl1A_E349S
(0.5 μM) showed a 45% yield (Table 3). This was in contrast to cases reported by
Moracci [13] on glycosynthases derived from thermophilic enzymes where reactions
performed at low temperature (35ºC) generated product yields lower than 10%,
suggesting that syntheses performed under these temperatures following the classical
approach are capable of producing good yields if experimental conditions are
optimized.
362
363
364
365
366
367
368
369
370
371
3.5 Glycosynthase activity using exogenous nucleophile
372
373
374
375
376
Another strategy that has been developed to conduct the glycosynthase reaction is in
situ generation of the activated glycosyl donor, which is particularly interesting for
thermophilic enzymes [13,27,28] as the conventional glycosyl fluoride donor is
unstable at high temperatures and requires the glycosynthase reaction to be carried out
at 30-37° C; a temperature far below the optimal activity of a thermophilic enzyme.
13
Page 13 of 32

377
378
379
380
The in situ generation of the α-glucosyl formate complex (by chemical rescue with
sodium formate, leading to a reactive adduct that mimics the glycosyl-enzyme
intermediate) creates a potential glycosyl donor for the glycosynthase reaction with
4NPGlc as the acceptor molecule (Fig.4).
381
382
383
384
385
386
387
388
Glycosynthase reactions were run using 6 mM 4NPGlc and 2 M sodium formate at
70° C, in 50 mM citrate-phosphate at the optimal pH 5.6. Further optimization of the
glycosynthase reaction was carried out by varying the sodium formate concentration
from 0.1 M to 3 M at different pH values (3 to 7), but the oligosaccharide synthetic
yields were lower at pHs above 6 (data not shown), and at low pH (3 to 4) no
glycosynthase reaction was detected at all. This indicated that the synthase reaction
performed by the TnBgl1A-glycosynthases shared the pH optimum of 5.6 with
hydrolytic reactions of the wt enzyme (Fig. 5 and Table 4).
389
390
391
When using the exogenous nucleophile approach, the enzyme was less regioselective
than in the previous approach that used α-glucosyl fluoride as the donor. The product
was mainly 4NP-β- laminaribioside, but 4NP-β- cellobioside was also detected in
lower amounts. Low concentrations (not quantifiable) of tri- and tetra-saccharides
were also detected by HPLC when the enzyme concentration was increased for the
reaction.
392
393
394
395
396
397
398
399
400
401
An observation made, during the glycosynthetic reaction using formate as the
exogenous donor molecule, was that glucose was formed by spontaneous hydrolysis
of the labile α-glucosyl formate intermediate. For this reason, it may be asserted that
this hydrolytic activity was in competition with the transglycosylation of the acceptor
molecule. Both reactions have been plotted in Figure 5, demonstrating that both the
hydrolytic activity and the transglycosylation reaction occur to approximately the
same extent for TnBgl1A_E349A and TnBgl1A_E349G.
402
403
404
405
406
407
408
In addition, it has been reported by Perugino and Trincone [29,30] that enhanced
activity of glycosynthases from a thermophilic origin can occur when using in situ
generation of the donor at low concentrations of exogenous nucleophile and low pH,
which was suggested to be caused by increased protonation of the acid/base that
facilitates an increased efficiency for the first step of the reaction (Fig. 4). However,
this was not observed for TnBgl1A glycosynthases and the synthetic activity
disappeared at lower pH (data not shown).
14
Page 14 of 32

409
410
411
3.6 Enzymatic glycosylation of flavonoids
A previous report showed that TnBgl1A has a high affinity for hydrolysing flavonoid
glucosides, with KM values of 0.06 and 0.13 mM for quercetin 4´-glucoside (Q4´Glc)
and quercetin 3-glucoside (Q3Glc) respectively; KM values lower than that of 4NPGlc
at 0.2 mM [26] provide a strong indication that these flavonoids could be used as
acceptors, thus expanding the use of TnBgl1A-glycosynthases towards antioxidant
modification. The benefit of modifying flavonoid structures is that their
physicochemical properties can be influenced, as well their bioavailability.
412
413
414
415
416
417
418
419
420
421
The results from the screening of flavonoid acceptors revealed that the synthesis of
glycosylated flavonoids was possible with TnBgl1A-glycosynthase. However, these
glycosynthases were shown to preferentially select a glycosylated flavonoid rather
than a non-glycosylated one as the acceptor molecule. The use of quercetin aglycone
as an acceptor did not result in any detectable glycosylation (data not shown), while
the use of Q3Glc was found to be more successful. The major product of the reactions
with Q3Glc was quercetin 3,4´diglucoside (Q3,4´diGlc), which was identified using
standards of this molecule by HPLC and Mass Spectrometry (MS). The yields
observed from the synthesis of di-glucosylated flavonoids were determined to be low
(Table 5) when compared to the amount of product formed in oligosaccharide
synthesis, holding true over both synthetic approaches (using either the GlcF or
4NPGlc with formate donor, respectively). The use of 2NPGlc as the donor molecule
resulted in a significant yield improvement (Table 5), which indicated that
accommodation at the -1 subsite (see also paragraph 3.7) provided better
intermolecular interactions, facilitating the glycosynthase activity.
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
The flavonoid reactions were carried out in presence of 50% (v/v) methanol, a solvent
that was added to dissolve the flavonoid. Optimization of the solvent concentration
may thus lead to increased yields, as in this case a co-solvent concentration was used
that has been observed to reduce activity of many enzymes. High co-solvent
concentration can however be more advantageous in combination with a thermostable
enzyme like TnBgl1A, as many thermostable enzymes are able to tolerate high solvent
concentrations [17, 31, 32].
440
441
15
Page 15 of 32

442
3.7 Structural considerations on glycosynthase reactions
443
444
445
446
447
448
449
450
451
Putative enzyme/substrate interactions in TnBgl1A were analysed by molecular
modelling to rationalize the observed regiospecificity in the glycosynthase reactions.
In addition, comparisons between the crystal structures of TnBgl1A_E349G (PDB-
code, to be added) and Bgl3_E383A (1GNX), an efficient glycosynthase previously
reported by Faijes [9], were carried out. Both enzymes were shown to be dimers in
their respective crystal structure, with a TIM (β/α)8 barrel fold and two conserved
motifs, TXNE and TENG, containing the conserved acid/base and nucleophile of the
GH1 family. In general, TnBgl1A_E349G was found to have an open active site with
more loops than Bgl3_E383A. The residues building the glycone (-1) subsite in GH1
are highly conserved interacting tightly with the donor molecule by hydrogen bonds
The residues building the glycone (-1) subsite in the GH1 enzymes are highly
conserved and interact tightly with the donor molecule by hydrogen bonding [3]. Both
enzymes were regioselective, as the linkage in the major glycosynthetic product of
both enzymes was found to be a β-1,3-glycosidic linkage. This would be influenced
by the accommodation of the donor and acceptor via tight interactions with the
residues at the -1 and +1 subsites, respectively. Important interacting residues of the -
1 subsite in TnBgl1A were determined to be W120, Y293, and the general base E164,
while W322 and W166 correspond to important residues at the +1 subsite (Fig 6A).
Similar interactions were observed in Bgl3; Figure 6b demonstrates the putative
important roles played by residues W148, Y311, and E292 in the -1 subsite, as well as
W370 in the +1 subsite. In both structures, the closer hydroxyl group of the acceptor
(Glc-pNP) to the anomeric carbon of the donor is the 3-hydroxyl (distances of 3.8-3.7
Å, respectively, Fig 6) in agreement with the observed β-1,3 specificity. At 70°C,
TnBgl1A-glycosynthases also produced a minor β-1,4-disaccharide product (using the
formate approach). Despite the longer distance between the anomeric carbon of the
donor and the 4-hydroxyl of the acceptor (about 6Å in the modelled structure in Fig
6a), formation of a β-1,4-glycosidic linkage may reflect the higher flexibility of the
enzyme at 70°C that results in less stringent substrate specificity.
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
TnBgl1A_E349G residues at the aglycone area (+2 and +3 subsites) interacting with
the Q3Glc were also identified, most of which were characterised as aromatic
hydrophobic residues: Trp322 displays stacking interaction with the flavonoid rings B
and C. Tyr175 is also well located to interact with the A-ring of the flavonoid, while
16
Page 16 of 32

475
476
477
478
E406 and H178 interact with the glucosyl moiety of the Q3Glc acceptor (Figure 7).
The 4´-hydroxyl of Q3Glc was determined to have a distance of 3.7 Å to the anomeric
carbon of GlcF and in this manner, the flavonoid appeared to fit as the acceptor
molecule at the TnBgl1A active site to form the Q3,4´-diGlc product.
479
480
4. Conclusion
481
482
483
484
485
486
487
488
The “classical” glycosynthase approach that utilized α-glycosyl fluoride donors was,
despite the lability of the fluoride donors and the necessity to use a sub-optimal
temperature of the thermophilic enzyme TnBgl1A, shown as a promising method for
oligosaccharide synthesis. Synthesis resulted in at least 60% yield and selectivity for
4NPLam formation. The yield by an exogenous nucleophile was higher than the yield
using in situ generation of the glycosyl donor. The approach using formate (as
exogenous nucleophile) also resulted in lower selectivity as both 4NPLam and
4NPCel were produced, albeit the latter at low concentrations.
489
490
491
In an attempt to explore new areas for the application of thermostable glycosynthases,
quercetin-3-glycoside was also used as an acceptor molecule for the glycosylation of
flavonoids. Although product yields were lower, production of quercetin-3,4´-
diglycoside was confirmed and this demonstrated the potential for the use of
glycosynthases in synthetic reactions that utilize a much broader range of acceptor
molecules.
492
493
494
495
496
Acknowledgements
497
498
499
500
501
We acknowledge the Swedish research council Formas (supporting the research
program SuReTech) and EU FP7 (supporting the project AMYLOMICS) for financial
support. Geneco Research School is highly appreciated for the travel grant that
supported Tania Pozzo as an exchange researcher at the University Ramon Llull,
Barcelona, Spain.
502
503
504
References
[1] C. S. Rye and S. G. Withers, Curr. Opin. Chem. Biol. 4, (2000) 573–580.
17
Page 17 of 32

505
506
[2] B. L. Cantarel, P. M. Coutinho, C. Rancurel, T. Bernard, V. Lombard, B.
Henrissat, Nucl. Acids Res. 37 (2009) D233–D238.
507
508
[3] S. R. Marana, IUBMB Life 58 (2008) 63–73.
[4] T. Desmet and W. Soetaert, Biocat. Biotrans. 29 (2011) 1–18.
509
510
[5] J. Thimm and J. Thiem, In: Glycoscience (eds B.O. Fraser-Reid, K. Tatsuta, J.
Thiem) Springer-Verlag Berlin Heidelberg (2008) 1387-1409.
511
512
513
[6] P. Monsan and F. Paul, FEMS Microbiol. Rev. 16 (2006) 187–192.
[7] D. H. Crout and G. Vic, Curr. Opin. Chem. Biol. (1998) 98–111.
[8] A. G. Watts and S. G Withers, Biochem. J. 380 (2004) e9-e10.
514
515
[9] M. Faijes, M. Saura-Valls, X. Perez, M. Conti, A. Planas, Carbohydr. Res. 341
(2006) 2055–2065.
516
[10] E. Andrés, H. Aragunde, A.Planas. Biochem. J. 458, (2014) 355-363.
517
518
[11] L. F. Mackenzie, Q. Wang, R. A. J. Warren, S. G. Withers, J. Am. Chem. Soc.
120 (1998) 5583–5584.
519
[12] C. Malet and A. Planas, FEBS Letters 440 (1998) 208–212.
520
521
[13] M. Moracci, A. Trincone, G. Perugino, M. Ciaramella, M. Rossi, Biochem. 37
(1998) 17262–17270.
522
523
[14] M. Vallmitjana, M. Ferrer-Navarro, R. Planell, M. Abel, C. Ausín, E. Querol, A.
Planas, J.-A. Pérez-Pons, Biochem. 40 (2001) 5975–5982.
524
525
[15] M. Juncosa, J. Pons, T. Dot, E. Querol, A. Planas, J. Biol. Chem. 269 (1994)
14530-14535.
526
527
[16] S. O. Ramchuran, O. Holst, E. Nordberg Karlsson, J. Biosci. Bioeng. 99 (2005)
477-84.
528
529
[17] P. Turner, D. Svensson, P. Adlercreutz, E. Nordberg Karlsson, J. Biotechnol. 130
(2007) 67-74.
530
531
[18] A.W. Schüttelkopf and D.M. Van Aalten, Acta Crystallogr. D60 (2004) 1355–
1363.
532
533
[19] B. Hess, C. Kutzner, D Van Der Spoel, E Lindahl, J. Chem. Theory Comput.
(2008) 435-447.
18
Page 18 of 32

534
535
[20] V. Hornak R. Abel, A. Okur, B. Strockbine, A. Roitberg, C. Simmerling,
Proteins, 15 (2006) 712-725.
536
537
[21] O. Trott, A. J. Olson, J. Comput. Chem. 31 (2010) 455-461.
[22] W. Humphrey, A. Dalke, K. Schulten, J. Mol. Graphics 14 (1996) 33-38.
538
539
[23] C. Turner, P. Turner, G. Jacobson, M Waldebäck, P Sjöberg, E. Nordberg
Karlsson, K. Markides. Green Chem 8 (2006) 949-959.
540
541
[24] S.J. Williams and S.G. Withers, Carbohydr. Res. 327 (2000) 27-46.
[25] M. Faijes, X. Pérez, O. Pérez, A. Planas, Biochem. 42 (2003) 13304–13318.
542
543
[26] S. Khan, T. Pozzo, M. Megyeri, S. Lindahl, A. Sundin, C. Turner, E. Nordberg
Karlsson, BMC Biochem. 12 (2011) 1-15.
544
545
[27] A. Trincone, G. Perugino, M. Rossi, M. Moracci, Bioorg. Med. Chem. Letters 10
(2000) 365–368.
546
547
[28] G. Perugino, A. Trincone, A. Giordano, J. van der Oost, T. Kaper, M. Rossi, M.
Moracci, Biochem. 42 (2003) 8484–8493.
548
549
[29] G. Perugino, P. Falcicchio, M. Michela Corsaro, I. Matsui, M. Parrilli, M. Rossi,
M. Moracci, Biocat. Biotrans. 24 (2006) 23–29.
550
551
[30] A. Trincone, A. Giordano, G. Perugino, M. Rossi, M. Moracci, ChemBioChem 6
(2005) 1431–1437.
552
[31] T. Hansson, and P. Adlercreutz, Biotechnol. Bioeng. 75 (2001) 656–665.
553
554
[32] T. Hansson, M. Andersson, E. Wehtje, P. Adlercreutz, Enz. Microb. Technol. 29
(2001) 527-534.
555
19
Page 19 of 32

555
556
Table 1
Primers used for site directed mutagenesis of TnBgl1A.
Primer
Sequence (5' to 3')
BglAE349GF
BglAE349GR
BglAE349AF
BglAE349AR
GAGGTGTACATCACAGGGAACGGAGCTGCATTC
GAATGCAGCTCCGTTCCCTGTGATGTACACCTC
GAGGTGTACATCACAGCGAACGGAGCTGCATTC
GAATGCAGCTCCGTTCGCTGTGATGTACACCTC
BglAE349SF GGAGGTGTACATCACATCGAACGGAGCTGCATTCG
BglAE349SR CGAATGCAGCTCCGTTCGATGTGATGTACACCTCC
557
558
Codons changed are underlined in bold.
559
560
20
Page 20 of 32

560
561
562
563
Table 2
Hydrolase activity determined at different temperatures (35°C and 80°C) and pH
values (5.6 and 7) for TnBgl1A and its nucleophile variants.
Vo/[E)] (s-1)
Enzyme
pH 5.6
80°C
pH 5.6
35°C
pH 7
80°C
pH 7
35°C
TnBgl1A
87
17
23
6
TnBgl1A_E349G
TnBgl1A_E349A
TnBgl1A_E349S
0.04
N.D.
0.20
0.04
0.03
0.13
0.24
0.02
1.2
0.01
N.D.
0.03
564
565
566
N.D = not detected. Reaction of 2mM of 4NPGlc and different enzyme concentrations
(0.01 to 0.5 μM)
567
568
21
Page 21 of 32

568
569
570
571
Table 3
Glucosynthetic evaluation of 4NPLam formation at 1 mM GlcF and 5 mM 4NPGlc
condition run at 35° C, pH 7, monitored using HPLC.
Enzyme
[E]
Vo
Vo/[E] Time Yield β-1,3
(%)
(μM) (M/s × 10-7)
(s-1)
(h)
15
15
TnBgl1A_E349G
TnBgl1A_E349G
TnBgl1A_E349A
TnBgl1A_E349A
TnBgl1A_E349S
Bgl3_E383A
0.5
2.5
0.5
1.7
0.5
0.05
0.05±0.0002
0.23±0.0010
N.D
0.009
0.009
29
62
0.04±0.003
0.07±0.049
4.14±0.566
0.002
0.013
8.299
15
15
21
45
60
0.5
572
573
N.D =Not Detected
574
575
22
Page 22 of 32

575
576
577
Table 4
Oligosaccharide synthesis using exogenous nucleophile.
Enzyme
pH
Conc
Vo
Vo/(E) Time
Yield
β1,3
[%]
Yield
β1,4
[%]
[μM] [M/s × 10-7]
[s-1]
[h]
TnBgl1A_
E349G
5.6
0.5
0.398±0.009 0.079
7
45
6
4
7
0.5
0.5
2.7
N.D
0.151±0.008 0.003
1.940±0.182 0.072
7
1
28
35
3
5.6
10
TnBgl1A_
E349A
5.6
0.5
0.272±0.047 0.054
7
42
13
4
7
0.5
0.5
2.5
N.D
0.074±0.001 0.015
0.371±0.006 0.014
7
1
15
16
3
4
5.6
TnBgl1A_
E349S
5.6
0.5
0.059±0.011 0.012
1
2
0.17
578
579
Reaction run with, 6 mM 4NPGlc and 2 M sodium formate, at 70° C product were
monitored by HPLC. N.D = Not Detected.
580
581
23
Page 23 of 32

581
582
583
Table 5
Enzymatic glycosylation of flavonoids.
Enzyme
Conc.
[μM]
Donor
1mM
Acceptor
3mM
Time
[h]
Yield
Q3,4-DiGlc [%]
TnBgl1A_
E349G
1
1
1
1
1
4NPGa
2NPGa
4NPGa
2NPGa
GlcF
Q3Glc
Q3Glc
Q3Glc
Q3Glc
Q3Glc
16
16
16
16
16
12
37
12
19
10
TnBgl1A_
E349A
TnBgl1A_
E349G
TnBgl1A_
E349A
1
GlcF
Q3Glc
16
6
584
585
586
Glycosynthase reaction using a ratio 1:3 donor acceptor, using both glycosynthetic
a
approaches, product monitored using HPLC. In situ donor formation using 2 M
sodium formate N.D = Not Detected
587
588
589
24
Page 24 of 32

*Graphical Abstract (for review)
THERMOPHILIC
GLYCOSYNTHASE
or
donor
acceptor
or
glycosylated product
or
Page 25 of 32

Figure 1
Figure 1.
●
pH5.6 80°C
●
●
pH7.0 80°C
pH5.6 35°C
●
pH7.0 35°C
0.01 0.02
0.05
0.08
Enzyme concentration [uM]
Fig.1. Linear dependence of initial velocity and enzyme concentration for TnBgl1A
hydrolase activity, determined at different conditions of pH (5.6 and 7) and temperature
(35°C and 80°C).
Page 26 of 32

Figure 2
Figure 2
Fig. 2. Chemical reactivation for TnBgl1A inactive nucleophile variants using different
concentrations of NaN3 at 35° C, pH 5.6 and a fixed enzyme concentration of 0.08 μM.
Page 27 of 32

Figure 3
Figure 3
Fig.3. Classical glycosynthase mechanism.
Page 28 of 32

Figure 4
Figure 4
Fig.4. Glycosynthase activity using formate as exogenous nucleophile
Page 29 of 32