Acceptor-induced modification of regioselectivity in CGTase-catalyzed glycosylations of p-nitrophenyl-glucopyranosides

Article abstract of DOI:10.1016/j.carres.2014.11.010

Cyclodextrin glycosyltransferases (CGTase) are reported to selectively catalyze α(1→4)-glycosyl transfer reactions besides showing low hydrolytic activity. Here, the effect of the anomeric configuration of the glycosyl acceptor on the regioselectivity of

Full text of DOI:10.1016/j.carres.2014.11.010

Carbohydrate Research 404 (2015) 46–54
Contents lists available at ScienceDirect
Carbohydrate Research
journal homepage: www.elsevier.com/locate/carres
Acceptor-induced modification of regioselectivity in CGTase-
catalyzed glycosylations of p-nitrophenyl-glucopyranosides
⇑
Simon Strompen, Alfonso Miranda-Molina, Agustín López-Munguía, Edmundo Castillo ,
Gloria Saab-Rincón
⇑
Departamento de Ingenería Celular y Biocatálisis, Instituto de Biotecnología, Universidad Nacional Autónoma de México, Apartado Postal 510-3, Cuernavaca, Morelos 62251, Mexico
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 30 June 2014
Received in revised form 30 October 2014
Accepted 17 November 2014
Available online 29 November 2014
Cyclodextrin glycosyltransferases (CGTase) are reported to selectively catalyze a(1?4)-glycosyl transfer
reactions besides showing low hydrolytic activity. Here, the effect of the anomeric configuration of the
glycosyl acceptor on the regioselectivity of CGTase catalyzed glycosylations was investigated. For this
purpose, the
lus macerans and Thermoanaerobacter sp. CGTases were used as biocatalysts and b-cyclodextrin as the gly-
cosyl donor. As expected, p-nitrophenyl- -glucopyranosyl-(1?4)-O- -glucopyranoside was
produced when p-nitrophenyl- -glucopyranoside was used as acceptor with B. macerans CGTase. Sur-
prisingly, when p-nitrophenyl-b- -glucopyranoside was used as glycosyl acceptor, besides the expected
(1?4)-glycosylation products both (1?3)- and (1?6)-transfer products were also obtained. This
unexpected change in B. macerans CGTase regioselectivity leading to (1?4)-, (1?3)- and (1?6)-gly-
a and b anomers of p-nitrophenyl-D-glucopyranoside were used as glycosyl acceptors, Bacil-
a
-
D
a-D
Keywords:
a
-D
Cyclodextrin glycosyltransferase (CGTase)
Glycosylation
Selectivity
p-Nitrophenyl-glucopyranoside
Anomeric configuration
D
a
a
a
a
a
a
cosylation products was also observed for Thermoanaerobacter sp. CGTase with the b anomer. It is shown,
applying time course analyses, that all isomers can be synthesized efficiently by adequate selection of
enzyme and reaction conditions. In fact, when using Thermoanaerobacter sp. CGTase the yield of p-nitro-
phenyl-b-
The previously unknown capacity of
D
-isomaltoside (the
a
(1?6)-transfer product) was the highest at long reaction time (19% yield).
(1?6)-glycosidic linkages formation by CGTases demonstrates an
a
unexpected broader regioselectivity of CGTases in glycosyl-transfer reactions as well as an acceptor
dependent transfer selectivity.
Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction
or vitamins.12 The hydrolytic activity (H2O as acceptor) of CGTases
has been reported to be negligible for most CGTases.
Cyclodextrin glycosyltransferases (CGTases, EC 2.4.1.19) cata-
lyze four types of reactions: (1) cyclization, (2) coupling, (3) dispro-
portionation, and (4) hydrolysis.1,2 The cyclization activity of
CGTases is used industrially for the intramolecular synthesis of
cyclic ring structures consisting of 6, 7, or 8 glucose units named
CGTases belong to family 13 of glycoside hydrolases, the large
group of -amylase family enzymes, which also include -amy-
lases, isoamylases, amylopullulanases, pullulanases, neopullulan-
ases, and branching enzymes.13 Despite their structural similarity
with other enzymes in the family catalyzing hydrolysis and/or gly-
a
a
a
-, b- and
c
-cyclodextrins, respectively, from starch or similar
cosyl-transfer of
have been reported to mainly catalyze
a
(1?4) and
a
(1?6)-glycosidic linkages, CGTases
(1?4)-glycosyl transfer
a(1?4)-linked glucose polymers. The coupling activity (i.e., the
a
opening of cyclodextrin rings with subsequent transfer to acceptor
molecules) and the disproportionation activity (transfer of linear
malto-oligosaccharides to suitable acceptors) of CGTases represent
a valuable tool for the intermolecular glycosylation of acceptor
molecules. Among them, sugars3,4 and glycosides4–6 have been
successfully glucosylated, but also non sugar-acceptors such as
sugar alcohols,7 polyols,8 polyphenols,9 flavonoides,10 saponins,11
reactions. Nevertheless, several examples can be found in the liter-
ature dealing with deviations from the characteristic CGTase regi-
oselectivity. In this context, the formation of glycosidic bonds other
than a(1?4) when glycosides or carbohydrates other than glucose
or its malto-oligosaccharides act as acceptors, has been reported.
For instance, the formation of neotrehalose by the glucosyl-trans-
fer to C1-OH of b-D-glucose has been observed for CGTases and
amylases, albeit to a much lesser extent as compared to the trans-
fer to C4-OH.14,15 Recently, the glucosylation of methyl- and
⇑
Corresponding authors. Tel.: +52 777 329 1640; fax: +52 777 317 2388.
E-mail addresses: edmundo@ibt.unam.mx (E. Castillo), gsaab@ibt.unam.mx
allyl-a-L-rhamnopyranoside has been studied by Urbach and
co-workers16 who found that both, the C-3 and C-4 positions of
(G. Saab-Rincón).
http://dx.doi.org/10.1016/j.carres.2014.11.010
0008-6215/Ó 2014 Elsevier Ltd. All rights reserved.

S. Strompen et al. / Carbohydrate Research 404 (2015) 46–54
47
the rhamnosyl-moiety of the acceptors, were glucosylated by Bacil-
lus circulans CGTase. Interestingly, the C4-OH-glucosylated product
was initially formed, representing the kinetic product, while long
reaction times thermodynamically favored the transfer of the glu-
cosyl-residues to the C-3 position of the rhamnosyl-acceptor.
Similarly, a remarkable regioselectivity, difficult to achieve by
non-enzymatic procedures has been demonstrated for CGTase-cat-
alyzed glycosylations of several flavonoids despite their multiple
hydroxyl-groups both in the aglycon and the sugar moieties. In
the case of naringin and neo-hesperidin, the glycosyl-transfer
was directed to position C-3 of the glucosyl-moiety using Bacillus
sp. CGTase.17 Rutin, genistin, and hesperidin on the other hand
were glycosylated at position C-4 of the glucosyl-moiety.18–20 Up
to date, no CGTase-catalyzed glycosylation of glucopyranosides to
positions other than C-1, C-3, or C-4 has been reported. Preliminary
results suggesting the formation of several glycoside-isomers from
the concentrations of transfer products with up to three glucose
units are shown in Figure 1b. Slow liberation of free p-nitrophenol
(<10% in the first 29 h) is also observed as a consequence of the
CGTase hydrolysis or the glycosyl-transfer from 2 to maltooligo-
saccharides as acceptors. The non-enzymatic hydrolysis of 2 was
found to be negligible under the applied reaction conditions. A
p-nitrophenyl-disaccharide was the major product obtained at an
analytical yield of 27%. As expected, NMR analysis confirmed the
a(1?4) regioselectivity of B. macerans CGTase. Indeed, the product
was identified as p-nitrophenyl- -glucopyranosyl-(1?4)-O-a-D-
a-D
glucopyranoside (4). No formation of other p-nitrophenyl-disac-
charide isomers was observed. The results suggest that the formed
p-nitrophenyl-oligosaccharides with more than two glucosyl-units
are most likely also linked by
In analogy, the reaction between the b-anomer p-nitrophenyl-
-glucopyranoside (3) as acceptor and b-cyclodextrin as donor
was investigated. As with the -anomeric acceptor substrate, a dis-
a(1?4) glucosidic bonds.
b-
D
CGTase-catalyzed glycosylations of p-nitrophenyl-b-
D-glucopyran-
a
oside prompted us to investigate the change of regioselectivity
induced by the anomeric configuration of the glycosyl-acceptor.
The anomeric control of regioselectivity has been described previ-
tribution of p-nitrophenyl-oligosaccharide products was obtained,
with a decrease in acceptor product concentration as the number
of glucosyl transferred residues increases, with p-nitrophenyl-pen-
taoside as the largest detectable product. An apparent equilibrium
in the synthesis of p-nitrophenyl-oligosaccharides is apparently
reached within 1 h under the applied reaction conditions (Fig. 2).
Unexpectedly, a further increase in p-nitrophenyl-disaccharides
concentration from 27 mM (5 h) to 38 mM (148 h) was observed
(Fig. 2, inset). These three p-nitrophenyl-b-disaccharide isomers
ously for galactosidase-catalyzed glycosylations using methyl
a- or
b-D glucoside as acceptor and p-nitrophenyl-b-N-acetyl-D-galactos-
aminide as donor.21 The present work, aims at an increased under-
standing of CGTase regioselectivity and to explore previously
unreported factors controlling it. For this purpose we studied the
glucosylation of the
ranoside with B. macerans and Thermoanaerobacter sp. CGTases.
The expected or b-p-nitrophenyl-maltooligosaccharides
obtained as a result of the regioselective CGTase glycosyl-transfer
toward the C4-OH glucose position in either the - or b-glucoside
represent useful substrates for the rapid detection of enzymes
bearing activity toward the corresponding or b glycosidic bonds.
The same type of glycosides have been successfully applied for the
a and b anomers of p-nitrophenyl-D-glucopy-
were identified as nitrophenyl-b-
glucopyranoside (p-nitrophenyl-b-
b- -glucopyranosyl-(1?3)-O- -glucopyranoside
phenyl-b- -nigeroside, 6), and p-nitrophenyl-b-
(1?6)-O- -glucopyranoside
7). This relaxed regioselectivity observed with the b-anomer when
D
-glucopyranosyl-(1?4)-O-
a-D-
a-
D-maltoside, 5), p-nitrophenyl-
D
a-D
(p-nitro-
a
D
D
-glucopyranosyl-
a-
D
(p-nitrophenyl-b-D-isomaltoside,
a
compared to the
below.
a-anomer as acceptor 2 is discussed in detail
detection and screening of
a-glucosidase, a-amylase or glyco-
synthase mutant activities.22–24
2.2. Structural analysis of disaccharide transfer products
2. Results and discussion
Disaccharyl glycosylation products of
a- and b-anomers of p-
nitrophenyl- -glucopyranoside (2, 3) were purified and subjected
to detailed structural analysis by 1D and 2D NMR spectroscopy.
D
2.1. B. macerans CGTase-catalyzed glycosylation of p-
nitrophenyl- /b- -glucopyranoside
a
D
Only a single disaccharide product (4) was observed in the B. mac-
erans CGTase glycosylation of the
tra of 4, two anomeric protons in
a
-anomer 2. In the 1H NMR spec-
The glycosylation of p-nitrophenyl-a/b-D-glucopyranoside was
carried out according to Scheme 1 using b-cyclodextrin as glyco-
syl-donor using B. macerans CGTase.
a
orientation were observed at d
5.66 (d, J = 3.6 Hz, H-1) and d 5.03 (d, J = 3.6 Hz, H-10). HSQC
showed their attachment to the anomeric carbons at dC 101.31
(C-10) and 97.81 (C-1), respectively. These signals characteristically
served as the starting point for the analysis of the 1H–1H COSY and
TOCSY experiments for connectivities within the spin system. The
inter glucosidic linkage was further confirmed from the HMBC
spectrum where a strong correlation was found between C-4 (d
The acceptor substrate p-nitrophenyl-a-D-glucopyranoside (2)
is consumed rapidly until an apparent equilibrium is reached at
45% conversion of 2 after 2 h using B. macerans CGTase (Fig. 1a
and b). p-Nitrophenyl-oligosaccharides with up to five glucosyl
units were formed in detectable amounts. It has been reported that
short molecular weight products are favored when a large acceptor
to donor substrate concentration ratio is used.25 For clarity, only
79.65) of the p-nitrophenyl-
a
-D
-glucopyranoside moiety of 4 and
OH
O
O
HO
O
OH
HO
1
O
OH
O
OH
HO
HO
O
OH
OH
OH
O
O
O
O
HO
OH
O
OH
OH
O
HO
HO
CGTase
HO
HO
OH
+
O
OH
O
OH
O
NO2
OH
HO
n
O
OH
OH
OH
O
NO2
O
OH
OH
OH
O
O
OH
2 or 3
HO
Scheme 1. Glycosyl-transfer reaction to p-nitrophenyl-
a-
D-glucopyranoside (2) or p-nitrophenyl-b-D-glucopyranoside (3) catalyzed by CGTase.

48
S. Strompen et al. / Carbohydrate Research 404 (2015) 46–54
Figure 1. B. macerans CGTase-catalyzed glycosylation of 2 (j, co = 50 mM) with b-cyclodextrin (1, co = 10 mM) as donor substrate at 40 °C. (a) Product distribution at
apparent equilibrium after 2 h. (b) Time course of reaction. p-Nitrophenyl- -trisaccharide (N), p-nitrophenol (}).
-maltoside (4), p-nitrophenyl-
a-
D
a
anomeric carbons at d 100.72 (C-10) and 99.50 (C-1). The carbon
to which each hydrogen was attached was defined from HSQC.
The proton-coupling network was established using a combination
of 1H–1H COSY, NOESY, TOCSY, HSQC and HMBC experiments.
Based on the HMBC correlations, the glucose moiety was linked
to C-4 of the substrate p-nitrophenyl-b-D-glucopyranoside (3) on
the basis of long-range correlation between H-4 (d 3.41) of accep-
tor and the C-10 of glucose moiety at d 100.72.
The 13C NMR spectrum of compound 6 contained 16 signals. In
the 1H NMR spectrum, two anomeric protons were observed:
d = 5.19 (d, J = 7.6 Hz, H-1) and 5.03 (d, J = 3.6 Hz, H-10), and HSQC
showed its attachment to the anomeric carbons at d 100.06 (C-10)
and 99.62 (C-1). This served as the starting point for tracing con-
nectivities within the sugar spin system by COSY and TOCSY exper-
iments. The 13C NMR spectrum of 6 gave two upfield resonances
which corresponded to methylenes (dC 60.66 and 60.19), while
the remaining resonances corresponded to methines. The carbon
Figure 2. Time course analysis of B. macerans CGTase-catalyzed glycosylation of p-
nitrophenyl-b- -glucopyranoside (co = 100 mM) (3) with b-cyclodextrin (1, co = 20 -
mM). p-Nitrophenyl-b-
p-nitrophenyl-b-trisaccharide (N), p-nitrophenyl-b-tetrasaccharide (}), p-nitro-
phenyl-b-pentasaccharide (.). The inset demonstrates the time course after
160 h reaction time.
D
D
-glucopyranoside 3 (j), p-nitrophenyl-b-disaccharide (4),
atom at d 85.57 (C-3) in the p-nitrophenyl-b-D-glucopyranoside-
substrate 3 was the point of linkage to glucose. The long-range cor-
relation between H-10 (d 5.03) of the glucose moiety and C-3 (d
85.57) of 3 in the HMBC spectrum indicated that the glucose resi-
H-10 at d 5.03 (d, J = 3.6 Hz) of glucose. This was confirmed by the
interactions observed between the anomeric hydrogen of glucose
at d 5.03 (H-10) and the proton at d 3.46 (H-4) of 2, in the NOESY
experiment. Therefore, the analysis of 4 allowed us to conclude
due is
a-linked to OH-3 of 3. Similar analysis by NOESY established
that the glucose residue was
a
-linked to O-3 of pNPbG because of
the correlation between H-10 of glucose at d 5.19 and H-3 of the
pNPbG at d 3.50. Therefore, compound 6 corresponds to p-nitro-
that the two glucopyranose rings were connected via
type linkage and that the glucose is -linked to C-4 of 2. According
to these results, the structure for the acceptor product
a-(1?4)
phenyl-b-
D-glucopyranosyl-(1?3)-O-
a-D
-glucopyranoside.
Full
a
assignments of the proton and carbon resonances were secured
from the TOCSY, COSY, NOESY and HMBC data.
4
In the 13C NMR spectra of 7, 16 carbon resonances were
observed, while 18 1H resonances in the 1H NMR spectra. The
chemical shifts for the 16 carbons were linked to those of the
hydrogens using HSQC. The characteristically downfield anomeric
carbons at d 100.29 (C-1) and 98.14 (C-10); and anomeric hydro-
gens at d 4.81 (J = 7.2 Hz, H-1) and 4.69 (J = 3.2 Hz, H-10) served
as the starting point for the analysis of the 1H–1H COSY for connec-
tivities within the spin system. 13C NMR spectra also showed sig-
nals at d 60.79 and d 66.41 indicated the presence of two
pyranoside type residues. The two diastereotopic H-6 hydrogens
at d 3.62 and at d 3.72 in the 1H spectrum were also useful markers
and attached to the carbon at d 66.41 in the 13C spectrum. The
other 13C peak at d 60.79 has its corresponding protons as a com-
plex multiplet at d 3.38 and d 3.51 and is hence assigned to non-
corresponds to p-nitrophenyl-a-D-glucopyranosyl-(1?4)-O-a-D-
glucopyranoside.
Surprisingly, analysis of the purified p-nitrophenyl-disaccharide
products of the b-anomer (3) revealed the formation of three
different disaccharide isomers in significant amounts. These
products were identified as p-nitrophenyl-b-
(1?4)-O- -glucopyranoside (p-nitrophenyl-b-
p-nitrophenyl-b- -glucopyranosyl-(1?3)-O- -glucopyranoside
(p-nitrophenyl-b- -nigeroside, 6), and p-nitrophenyl-b- -gluco-
pyranosyl-(1?6)-O- -glucopyranoside (p-nitrophenyl-b-
isomaltoside, 7) (Scheme 2).
D-glucopyranosyl-
a
-D
D-maltoside, 5),
D
a-D
D
D
a-
D
D-
The 1H NMR spectra of 5, 6, and 7 indicated the presence of a set
of signals of an aromatic AB system [d 8.22 (2H, d, J = 9.2 Hz, Ar);
7.22 (2H, d, J = 9.2 Hz, Ar)], corresponding to the p-nitrophenyl res-
idue. A detailed analysis of the unassigned NMR correlations in the
2D NMR spectra revealed that the principal transfer product was
linked C-6. Based on the HMBC correlations, the
a-glucose was
positioned at C-6 of the 3, which was confirmed by three-bond cor-
relation of their two CH2-O protons nonequivalent (d 3.62 and
3.72) to carbon (C-10) of glucose at d 98.14. Moreover, HMBC exper-
iment also showed a strong correlation between H-10 (d 4.69) of
glucose and C-6 (d 66.41) of 3 further confirming the position of
the p-nitrophenyl-b-D-glucopyranosyl-(1?4)-O-a-D-glucopyrano-
side 5, and the presence of the glucopyranosyl units was deduced
from the 1H NMR anomeric protons at d 5.16 (d, J = 8.0 Hz, H-1)
and d 5.06 (d, J = 4.0 Hz, H-10), which were compatible with the

S. Strompen et al. / Carbohydrate Research 404 (2015) 46–54
49
OH
OH
OH
O
O
HO
O
O
OH
HO
HO
HO
1
O
O
HO
HO
OH
OH
OH
O
OH
OH
HO
HO
O
O
O
OH
OH
HO
O
OH
O
O
O
HO
O
O
CGTase
NO2
NO2
OH
O
OH
O
OH
OH
HO
HO
HO
O
OH
+
NO2
5
O
OH
6
OH
OH
OH
OH
O
O
OH
OH
O
O
HO
HO
O
OH
3
HO
OH
OH
O
O
HO
HO
O
NO2
OH
7
Scheme 2. Glucosyl-transfer reaction to p-nitrophenyl-b-D-glucopyranoside (3) catalyzed by CGTase (formed p-nitrophenyl-maltooligosaccharides not shown).
the glycosylation. According to these results, the structure for the
compound may be established as p-nitrophenyl-b- -glucopyrano-
syl-(1?6)-O- -glucopyranoside (7). Compound 7 contains a
bound C-6 with a chemical shift at d 66.41, which is significantly
farther downfield that the free C-6 at d 60.79. It has been shown
that glycoside bond formation causes the chemical shift of the car-
bon atom involved to be displaced downfield. The structures of
compounds 6 and 7 are reported here for the first time.
of the a(1?4)-glycosidic linkage may be possible. The latter mech-
D
anism, however, is less likely. HPLC analyses did not indicate the
formation of significant quantities of required p-nitrophenyl-
maltotrioside isomers as intermediate products. Additionally, com-
a
-D
pound 5 is a potential acceptor as it contains an
a-1,4-glycosidic
linkage that resembles the natural oligosaccharide acceptors, for
which no glycosylation in position C-3 or C-6 has been reported.
The effect of the p-nitrophenyl aglycon is expected to be more pro-
nounced for p-nitrophenyl-b-D-glucopyranoside as acceptor, as the
2.3. Time course of glucosylation of p-nitrophenyl-b-
glucopyranoside
D
-
p-nitrophenyl group is in closer proximity to the active site, as
observed in preliminary docking simulations (see later).
It should be noted, that under the given reaction conditions,
particularly the long reaction times, the enzyme activity may have
severely decreased. After 50 h at 40 °C in the same reaction buffer
in the absence of substrates, B. macerans CGTase activity was found
to decrease 93% from the initial activity (Fig. S1, Supporting
information). However, it must be considered that the presence
of substrates can have a stabilizing effect.16 Additionally, similar
quantities of disaccharide isomers were observed in a control
experiment where enzyme activity available was maintained
through fed batch addition of B. macerans CGTase to compensate
for the loss of activity (Fig. S2, Supporting information).
A time course analysis of the p-nitrophenyl disaccharide iso-
mers formed according to Scheme 2 was carried out. It was found
that using B. macerans CGTase as biocatalyst, mainly p-nitro-
phenyl-b-
favored product (Fig. 3a). This observation is expected considering
the CGTase specificity to form (1?4)-glycosidic linkages, as
reported for most glucopyranoside-acceptors studied.7 The forma-
tion of p-nitrophenyl-b- -nigeroside (6) occurs more slowly, but is
D-maltoside (5) is formed initially as the kinetically
a
D
already detectable after 5 h of reaction. At long reaction times, the
yield of 5 decreases while the concentration of 6 increases until
approximately the same level of 5 is reached at 18 mM final con-
centration (Fig. 3b). Additionally, a slow formation of up to
2.4. Glycosylation of p-nitrophenyl-a/b-D-glucopyranoside using
3.4 mM p-nitrophenyl-b-
148 h of reaction. Mechanistically, both a direct transfer of b-cyclo-
dextrin or maltooligosaccharides to p-nitrophenyl-b- -glucopyran-
oside as well as the formation of p-nitrophenyl-b- -maltoside (5)
or longer chain p-nitrophenyl-b- -maltooligosaccharides with sub-
sequent glycosylation at position C-3 or C-6 followed by hydrolysis
D-isomaltoside (7) was observed after
Thermoanaerobacter sp. CGTase
D
In order to evaluate if the observed regioselectivity is a general
property of CGTases or related to unusual characteristics of the B.
macerans enzyme, the regioselectivity of a CGTase of a different
D
D
origin was explored. Therefore the glucosylation of the
a- and
Figure 3. B. macerans CGTase-catalyzed glycosylation of 4. (a) Initial time course. (b) Full time course. p-Nitrophenyl-b-
D-maltoside (5, j), p-nitrophenyl-b-
D-nigeroside (6,
4), p-nitrophenyl-b-
D-isomaltoside (7, N). For clarity, concentrations of the p-nitrophenyl-b-
D-glucopyranoside and larger p-nitrophenyl-oligosaccharides have been omitted
in the graph.

50
S. Strompen et al. / Carbohydrate Research 404 (2015) 46–54
b-anomers of p-nitrophenyl-
D-glucopyranoside with Thermoanae-
a third product. Interestingly, long reaction times result in a
robacter sp. CGTase was also studied.
With p-nitrophenyl- -glucopyranoside (2) as acceptor, simi-
lar concentrations of p-nitrophenyl- -maltooligosaccharides were
obtained initially as shown in Figure 4a. However, product yields
after 1 h of reaction decreased significantly due to the rapid liber-
ation of p-nitrophenol making Thermoanaerobacter sp. CGTase less
decrease of concentrations not only of 5 but also of 6, after reach-
ing maxima of 25 and 18 mM, respectively. Additionally, the prod-
uct 7, that is, the disaccharide containing a (1?6)-glycosidic
linkage, does not only represent a minor by-product but the ther-
modynamically favored reaction product, obtained at 19% yield
(19 mM) after 148 h of reaction time (Fig. 5b).
a-D
a
suited for application in the synthesis of p-nitrophenyl-
igosaccharides. As outlined above, CGTases may cleave the
a
-maltool-
-gly-
a
2.5. Comparison of selectivity using Thermoanaerobacter sp.
and B. macerans CGTase
cosidic linkage to the p-nitrophenyl aglycon and liberate p-
nitrophenol through hydrolysis or transglycosylation depending
on the acceptor of the glucose molecule. Therefore, a structural
analysis of disaccharide products was not possible. However, based
on the general preference of CGTases toward the formation of
CGTase-catalyzed glycosylation of 3 as discussed above using
Thermoanaerobacter sp. and B. macerans CGTase was performed
using the same initial enzyme activity of both enzymes. CGTase
activity measurement is, however, based on the enzyme’s ability
to catalyze b-cyclodextrin formation from starch. This property is
not necessarily related or proportional to the transferase activity,
particularly toward the acceptors here studied. As a consequence,
equal enzyme activity expressed as cyclodextrin formation from
starch is not necessarily equivalent to equal activity glycosylation
of 3. In theory, the differences in progress curves observed for
the two CGTases result from differences in affinity and catalytic
activity of the two enzymes toward the alternative substrate.
a
(1?4)-glycosidic bonds as observed also for B. macerans CGTase,
it may be inferred that Thermoanaerobacter sp. CGTase also cata-
lyzed (1?4) glycosylation of 2. Reactions with Thermoanaerobact-
a
er sp. CGTase were carried out at 50 °C, while those of B. macerans
CGTase at 40 °C. However, control reactions with Thermoanaerob-
acter sp. CGTase at 40 °C followed a virtually identical course of
the reaction at a slowed down rate.
In contrast Thermoanaerobacter CGTase did not cleave substrate
3 in significant amounts to liberate p-nitrophenol under the
applied reaction conditions even after long reaction times. p-Nitro-
phenyl-b-maltooligosaccharides with up to five glucosyl units
were formed initially in detectable amounts as found previously
with B. macerans CGTase. However, longer reaction times led to a
reduction of the p-nitrophenyl-oligosaccharides concentration
and an increase in the glucopyranoside substrate (Fig. 4b). A max-
imum substrate conversion of 52% was obtained after 2 h. After
approximately 50 h, when the reaction reached an apparent
equilibrium, p-nitrophenyl-oligosaccharides with three or
more glucosyl-units disappeared leaving only the substrate
p-nitrophenyl-glucopyranoside and its monoglucosylated product.
We can define CGTase selectivity toward p-nitrophenyl-b-D-
maltoside 5 formation as the ratio of the concentration of 5 and
the sum of the concentration of all formed disaccharides (5, 6,
and 7) at any given time. A plot of the selectivity defined in these
terms vs. conversion of 3 clearly shows in a time-independent
manner the higher regioselectivity of B. macerans CGTase (Fig. 6).
It is found that in reactions with B. macerans CGTase, the disaccha-
ride 5 can be synthesized exclusively up to about 40% conversion of
3, without the synthesis of any of the other disaccharide-isomers
eventually formed. Higher degrees of conversion lead to a rapid
decrease of selectivity. Nevertheless, the selectivity toward 5 drops
at lower substrate conversion with the Thermoanaerobacter sp.
enzyme: in particular to 0.93 at a conversion of 0.4 when the
enzyme from B. macerans is still exclusively synthesizing p-nitro-
It is hypothesized that a hydrolytic side reaction of a(1?4)-glyco-
sidic bonds shifts the equilibrium toward the formation of mono-
and small oligosaccharides.
Remarkably, increased amounts of the same three disaccharide
isomers obtained with B. macerans CGTase were found. As in the
case of the B. macerans enzyme, p-nitrophenyl-b-D-maltoside (5)
also represents the kinetically controlled product of Thermoanae-
robacter sp. CGTase glycosylation of 3 using b-cyclodextrin (1) as
donor (Fig. 5a). After 0.5 h of reaction, the yield of 5 decreases, con-
phenyl-b-D-maltoside. The latter CGTase is therefore better suited
for the kinetically controlled production of this compound.
High concentrations of 6 can also be obtained using either
CGTase. The purification protocol involves selective glucoamylase
digestion of
graphic separation from the substrate p-nitrophenyl-b-glucopy-
ranoside and the product p-nitrophenyl-b- -isomaltoside 7. Thus,
for the preparative synthesis of 6 only the maximal concentration
a(1?4) glycosidic linkages and subsequent chromato-
comitant to the formation of p-nitrophenyl-b-
D-nigeroside (6), a
D
second isomer product, and p-nitrophenyl-b- -isomaltoside (7) as
D
Figure 4. Thermoanaerobacter sp. CGTase-catalyzed glycosylation of (a) p-nitrophenyl-
conditions: p-nitrophenyl- -glucopyranoside 2 (j, co = 50 mM), b-cyclodextrin (1, co = 10 mM), 50 °C. p-Nitrophenyl-
-glucopyranoside (j, co = 100 mM), b-cyclodextrin (1, co = 20 mM), 50 °C. p-Nitrophenyl-b-D-
a-
D-glucopyranoside 2 and (b) p-nitrophenyl-b-
D-glucopyranoside 3. (a) Reaction
a
-
D
a
-
D
-disaccharide (4), p-nitrophenyl-
a
-trisaccharide
(N), p-nitrophenol (}). (b) Reaction conditions: p-nitrophenyl-b-
D
3
glucopyranoside 4 (j), p-nitrophenyl-b-disaccharide (4), p-nitrophenyl-b-trisaccharide (N), p-nitrophenyl-b-tetrasaccharide (}), p-nitrophenyl-b-pentasaccharide (.).
The inset demonstrates the initial time course of the reaction.

S. Strompen et al. / Carbohydrate Research 404 (2015) 46–54
51
Figure 5. Thermoanaerobacter sp. CGTase-catalyzed glycosylation of 4. (a) Initial time course. (b) Full time course. p-Nitrophenyl-b-D-maltoside (5, j), p-nitrophenyl-b-D-
nigeroside (6, 4), p-nitrophenyl-b-
D-isomaltoside (7, N). For clarity, concentrations of the pnitrophenyl-b-D-glucopyranoside and larger p-nitrophenyl-oligosaccharides have
been omitted in the graph.
CGTase E257Q/D229N mutant, that
c-cyclodextrin binds to the
enzyme in a manner different from that of the linear maltooligo-
saccharides.37 Particularly, contacts with Phe183 and Phe195
responsible for positioning the substrate in subsite +2 by stacking
interactions and Tyr195 forming hydrogen bonds with OH-6 and/
or O5 groups of the sugar acceptor in subsite +1 were altered sig-
nificantly for c-cyclodextrin and the linear maltooligosaccharides,
respectively. Thus, the sugar acceptor can be accommodated in
the active site with some flexibility. It can therefore be inferred
that the aglycon and its anomeric orientation influence the bind-
ing of the glucosyl residue at subsite +1 leading to the formation
of several regioisomers as observed in this study. Preliminary
observation in docking experiments with the p-nitrophenyl-a-
glucoside in the active site of the T. thermosulfurigenes CGTase
shows that although the p-nitrophenyl group does not occupy
the exact position of the G+2 of acarbose, it fits well into the
active site. On the other hand, when the p-nitrophenyl-b-gluco-
side is placed, aligning the glucoside with the G+1 from acarbose,
the p-nitrophenyl group clashes with Tyr101 in the protein. If C-6
of the glucoside group is aligned with C-4 from G+1 in the acar-
bose, then the p-nitrophenyl group orients more similarly to G+2
from acarbose explaining the different specificities observed
experimentally in this work for the two anomers (Fig. S9, Sup-
porting information). Nevertheless, more detailed docking studies
are necessary to demonstrate the extent of the interactions in dif-
ferent enzymes such as B. macerans and Thermoanaerobacter sp.
CGTase, that are beyond the scope of this work.
Figure 6. Selectivity of CGTase catalyzed glycosylation of 4 toward formation of p-
nitrophenyl-b- -maltoside (5) expressed as 5/(5 + 6 + 7) as a function of conversion.
B. macerans CGTase (N). Thermoanaerobacter sp. CGTase (4).
D
of the compound is important. Impurities of other maltooligosac-
charides including disaccharide isomers are removed during the
process. The thermodynamically controlled synthesis of p-nitro-
phenyl-b-D-isomaltoside 7 can be carried out effectively using
Thermoanaerobacter sp. CGTase with up to 19% yield after 148 h,
while only 3.4% yield can be obtained with B. macerans CGTase
after 148 h even with fed batch addition of the enzyme to compen-
sate for activity loss (Fig. S2, Supporting information). The observa-
tion further underlines the differences in regioselectivity between
both CGTases.
3. Conclusion
2.6. Structural considerations
It is shown that the well known CGTases from B. macerans and
Thermoanaerobacter sp. frequently applied in various type of glu-
cosylation reactions display a broader than expected regioselectiv-
CGTases from different organisms that have been resolved in
free and complexed form.26–37 The 3D structure of Thermoanae-
robacter sp. CGTase (Toruzyme) is not known. However, the crys-
tal structure of B. macerans CGTase has been deposited in the
Protein Data Bank recently with code 4cjl. Its active site is very
similar to the one from Thermoanaerobacter thermosulfurigenes,
crystallized with acarbose (1a47),38 From these data, at least
ten sugar binding subsites, labeled ꢀ7 to +3, have been identified
in CGTases with active site residues Asp229 and Glu257 being
located between subsites ꢀ1 and +1.10 The acceptor substrate
can be accommodated at subsites +1 to +3. The glucosyl moiety
of the acceptor p-nitrophenyl-glucopyranoside, the substrate used
in this study, binds at subsite +1 while its p-nitrophenyl group
will primarily interact with the enzyme at subsite +2. Uitdehaag
and coworkers demonstrated for the Bacillus circulans strain 251
ity in the
formation of
were formed using p-nitrophenyl-b-glucopyranoside as acceptor
of glucose transferred from b-cyclodextrin, with the (1?4)-trans-
fer being the kinetically preferred reaction, resulting in the
synthesis of p-nitrophenyl-b- -maltoside, and p-nitrophenyl-b-
isomaltoside resulting from (1?6)-glycosylation as the thermo-
dynamically controlled reaction product. With the corresponding
-anomer, p-nitrophenyl- -glucopyranoside as acceptor, the for-
mation of (1?4)-glycosidic linkages was exclusively observed,
a-glycosylation of glucopyranosides. In particular, the
a
(1?4)-, (1?3)- and (1?6)-glycosidic linkages
a
a
a
D
D-
a
a
a
a
indicating the importance of the anomeric configuration of the
aglycon in position C-1. These results contribute to the under-
standing of side products formation observed previously by other

52
S. Strompen et al. / Carbohydrate Research 404 (2015) 46–54
groups9,21 and indicate the importance of analyzing whole
progress curves rather than conversions after short incubation
times. In fact, the promiscuous selectivity of CGTases observed
here as a side activity, can actually be used in the regioselective
glycosylation synthesis at three different positions of the glucosyl
moiety of the acceptor, as demonstrated here. Our results as well
as the aforementioned studies reported in the literature stress
the importance of the anomeric configuration of the acceptor, not
limited to p-nitrophenyl. More research is required in order to
determine factors governing the positional selectivity of CGTases
such as anomeric configuration and nature of the aglycon or other
neighboring functional groups.
4.2.2. 1H and 13C NMR analysis of transfer products
1H and 13C NMR was recorded on a Varian Unity NMR Spec-
trometer operating at 400 MHz in DMSOd6. Chemical shifts (d)
were quoted in parts per million (ppm). Coupling constant (J)
was reported in Hertz (Hz). Interglycosidic connectivity of the glu-
cosylation products was confirmed by both 1D and standard
homo-(COSY, TOCSY and NOESY) and heteronuclear (HSQC and
HMBC) correlated two-dimensional (2D) (Figs. S10–S28, Support-
ing information).
4.2.3. p-Nitrophenyl-a-D-glucopyranosyl-(1?4)-O-a-D-
glucopyranoside (4)
White amorphous powder (21% isolated yield); 1H NMR
(400 MHz, DMSOd6): d = 8.22 (d, 2H, J = 9.2 Hz, Ar), d = 7.22 (d,
2H, J = 9.2 Hz, Ar), d = 5.66 (d, 1H, J1,2 = 3.6 Hz, H-1), 5.03 (d, 1H,
4. Experimental
4.1. Material
J1 ,2 = 3.6 Hz, H-10), 3.89 (dd, 1H, J2,3 = 8.4 Hz, J3,4 = 9.2 Hz, H-3),
3.60 (m, 2H, H-6a0, H-6a), 3.50 (dd, 1H, J1,2 = 3.6 Hz, J2,3 = 10.0 Hz,
H-2), 3.46 (t, 1H, J3,2 = 7.2 Hz, J3,4 = 7.6 Hz, H-4), 3.42 (m, 1H, H-
5), 3.41 (m, 1H, H-50), 3.40 (m, 2H, H-6b0, H-6b), 3.38 (t, 1H,
0
0
p-Nitrophenyl-
a
-D
-glucopyranoside, p-nitrophenyl-b-
D-gluco-
pyranoside, p-nitrophenyl-
a-D
-maltoside and p-nitrophenyl-b-D-
J2 ,3 = 8.8 Hz, J3 ,4 = 9.2 Hz, H-30), 3.24 (dd, 1H, J1 ,2 = 3.6 Hz,
0
0
0
0
0
0
maltoside were purchased from Sigma Aldrich (St. Louis, MO,
USA). b-Cyclodextrin was obtained from American Maize-Products
Company (Hammond, IN, USA). B. macerans CGTase (‘Amano’-K)
was kindly provided by Amano Enzyme Inc. (Nagoya, Japan). Ther-
moanaerobacter sp. CGTase (Toruzyme 3.0L) was provided by
Novozymes A/S (Bagsvaerd, Denmark). Aspergillus niger glucoamy-
lase (GA) was obtained from EDC as Enzeco Glucoamylase Powder
AN (New York, NY, USA).
J2 ,3 = 9.6 Hz, H-20), 3.05 (dd, 1H, J3 ,4 = 8.8 Hz, J3 ,5 = 9.2 Hz, H-40).
13C NMR (100 MHz, DMSOd6): d = 162.49 (qc, C-O, Ar), 142.07 (qc,
C-NO2, Ar), 126.16 (CH ꢁ2, Ar), 117.33 (CH ꢁ2, Ar), 101.31 (C-10),
97.81 (C-1), 79.65 (C-4), 73.83 (C-50), 73.72 (C-30), 73.17 (C-3),
72.95 (C-5, C-20), 71.24 (C-2), 70.33 (C-40), 61.22 (C-60), 60.47 (C-6).
0
0
0
0
0
0
4.2.4. p-Nitrophenyl-b-D-glucopyranosyl-(1?4)-O-a-D-
glucopyranoside (5)
White amorphous powder (23% isolated yield); 1H NMR
(400 MHz, DMSOd6): d = 8.22 (d, 2H, J = 9.2 Hz, Ar), d = 7.22 (d,
2H, J = 9.2 Hz, Ar), d = 5.16 (d, 1H, J1,2 = 8.0 Hz, H-1), 5.06 (d, 1H,
4.2. Analytical methods
4.2.1. HPLC analysis
J1 ,2 = 4.0 Hz, H-10), 3.72 (m, 1H, H-6b0), 3.64 (m, 1H, H-6b), 3.63
0
0
(m, 1H, H-6a0), 3.59 (m, 1H, H-50), 3.58 (m, 1H, H-5), 3.48 (dd,
The separation of p-nitrophenyl-a/b-maltooligosaccharides for
1H, J2 ,3 = 7.6 Hz, J3 ,4 = 8.8 Hz, H-30), 3.47 (m, 1H, H-6a), 3.41 (dd,
1H, J3,2 = 9.2 Hz, J3,4 = 9.2 Hz, H-4), 3.39 (dd, 1H, J2,3 = 9.2 Hz,
quantification was carried out using a Prevail Carbohydrate ES 5u
column (4.6 ꢁ 250 mm, Grace Davison, Deerfield, IL, USA) with
acetonitrile (ACN): water 75:25 as a mobile phase. Quantification
0
0
0
0
J3,4 = 9.6 Hz, H-3), 3.25 (dd, 1H, J1 ,2 = 3.6 Hz, J2 ,3 = 9.6 Hz, H-20),
0
0
0
0
3.07 (dd, 1H, J3 ,4 = 8.8 Hz, J3 ,5 = 9.2 Hz, H-40). 13C NMR
(100 MHz, DMSOd6): d = 162.26 (qc, C-O, Ar), 141.73 (qc, C-NO2,
Ar), 125.72 (CH ꢁ2, Ar), 116.53 (CH ꢁ2, Ar), 100.72 (C-10), 99.50
(C-1), 78.85 (C-4), 76.02 (C-50), 75.41 (C-5), 73.56 (C-30), 73.30
(C-3), 72.64 (C-2), 72.42 (C-20), 69.87 (C-40), 60.81 (C-6), 60.06
(C-60).
of nitrophenyl-
detection at 300 nm, respectively. No separation of p-nitro-
phenyl-b- -glucopyranosyl-(1?3)-O- -glucopyranoside (p-
nitrophenyl-b- -nigeroside) and p-nitrophenyl-b- -glucopyrano-
syl-(1?4)-O- -glucopyranoside (p-nitrophenyl-b- -maltoside)
could be achieved on a Prevail carbohydrate ES 5u column, while
p-nitrophenyl-b- -glucopyranosyl-(1?6)-O- -glucopyranoside
(p-nitrophenyl-b- -isomaltoside) could be separated from the
(1?3) and (1?4)-isomers (Fig. S3, Supporting information).
Calibration for quantification was performed with the substrate
p-nitrophenyl- /b-glucopyranoside as well as all pure p-nitro-
phenyl-disaccharide isomers ( (1?3)-; (1?4)-; (1?6)). Only
a
/b-maltooligosaccharides was achieved using UV
0
0
0
0
D
a-D
D
D
a
-D
D
D
a-D
D
4.2.5. p-Nitrophenyl-b-D-glucopyranosyl-(1?3)-O-a-D-
a
a
glucopyranoside (6)
White amorphous powder (13% isolated yield); 1H NMR
(400 MHz, DMSOd6): d = 8.22 (d, 2H, J = 9.2 Hz, Ar), d = 7.22 (d, 2H,
J = 9.2 Hz, Ar), d = 5.19 (d, 1H, J1,2 = 7.6 Hz, H-1), 5.03 (d, 1H,
a
a
a
a
J1 ,2 = 3.6 Hz, H-10), 3.76 (ddd, 1H, J5 ,6a = 2.0 Hz, J5 ,6b = 5.8 Hz,
negligible differences were found for the signal intensities of dif-
ferent isomers or p-nitrophenyl-maltooligosaccharides of different
chain lengths, presumingly due to the absorption stemming solely
0
0
0
0
0
0
J4 ,5 = 10.2 Hz, H-50), 3.69 (dd, 1H, J5,6b = 5.4 Hz, J6a,6b = 10.6 Hz,
H-6b), 3.63 (m, 1H, H-6a0), 3.52 (m, 1H, H-5), 3.51 (m, 2H, H-4,
H-6a), 3.50 (dd, 1H, J2,3 = 8.4 Hz, J3,4 = 8.8 Hz, H-3), 3.49 (m, 1H, H-
0
0
from the aglycon. The p-nitrophenyl-b-
D-nigeroside (a(1?3)) and
40), 3.47 (m, 2H, H-2, H-6b0), 3.26 (dd, 1H, J1 ,2 = 3.6 Hz, J2 ,3 = 7.2 Hz,
p-nitrophenyl-b- -maltoside ( (1?4)) could be separated on a
D
a
0
0
0
0
H-20), 3.13 (dd, 1H, J2 ,3 = 9.2 Hz, J3 ,4 = 9.6 Hz, H-30). 13C NMR
(100 MHz, DMSOd6): d = 162.24 (qc, C-O, Ar), 141.77 (qc, C-NO2,
Ar), 125.75 (CH ꢁ2, Ar), 116.58 (CH ꢁ2, Ar), 100.06 (C-10), 99.62
(C-1), 85.57 (C-3), 76.52 (C-4), 73.50 (C-2), 72.76 (C-50), 72.48 (C-
20), 71.64 (C-5), 70.03 (C-30), 68.92 (C-40), 60.66 (C-60), 60.19 (C-6).
CarboPac PA200 column (3 ꢁ 250 mm, Dionex, Sunnyvale, CA,
USA) using a 98.5% of 150 mM NaOH (A) and 1.5% of 150 mM
NaOH, 500 mM NaOAc (B) as mobile phase (Fig. S4, Supporting
information). An ED50 Electrochemical Detector (Dionex,
Sunnyvale, CA, USA) was used for detection. Calibration was car-
0
0
0
0
ried out using the two pure isomers p-nitrophenyl-b-
and p-nitrophenyl-b- -maltoside. Quantification of p-nitro-
phenyl-b- -nigeroside and p-nitrophenyl-b- -maltoside was
therefore initially determined as a mixture by UV detection. Subse-
quent determination of the ratio of both isomers was then used to
calculate the respective concentration of each isomer. Compounds
not containing p-nitrophenol, that is, b-cyclodextrin or formed oli-
gosaccharides, were not quantified in this study.
D-nigeroside
D
4.2.6. p-Nitrophenyl-b-D-glucopyranosyl-(1?6)-O-a-D-
D
D
glucopyranoside (7)
White amorphous powder (5% isolated yield); 1H NMR
(400 MHz, DMSOd6): d = 8.22 (d, 2H, J = 9.2 Hz, Ar), d = 7.22 (d,
2H, J = 9.2 Hz, Ar), d = 4.81 (d, 1H, J1,2 = 7.2 Hz, H-1), 4.69 (d, 1H,
J1 ,2 = 3.2 Hz, H-10), 3.72 (m, 1H, H-6a), 3.63 (dd, 1H, J2,3 = 8.8 Hz,
0
0
0
0
J3,4 = 10.0 Hz, H-3), 3.62 (m, 1H, H-6b), 3.51 (dd, 1H, J5 ,6a = 5.0 Hz,

S. Strompen et al. / Carbohydrate Research 404 (2015) 46–54
53
J6a ,6b = 9.4 Hz, H-6a0), 3.43 (m, 1H, H-4), 3.42 (m, 1H, H-30), 3.38
zymogram analysis result from CGTase degradation (Fig. S8b, Sup-
porting information).
0
0
(m, 1H, H-6b0), 3.33 (m, 1H, H-2), 3.22 (dd, 1H, J1 ,2 = 4.0 Hz,
0
0
J2 ,3 = 9.2 Hz, H-20), 3.17 (m, 1H, H-5), 3.07 (ddd, 1H, J5 ,6a = 4.8 Hz,
0
0
0
0
J5 ,6b = 8.8 Hz, J4 ,5 = 8.8 Hz, H-50), 3.04 (m, 1H, H-40). 13C NMR
(100 MHz, DMSOd6): d = 162.45 (qc, C-O, Ar), 141.84 (qc, C-NO2,
Ar), 125.89 (CH ꢁ2, Ar), 116.90 (CH ꢁ2, Ar), 100.29 (C-1), 98.14
(C-10), 76.56 (C-40), 74.97 (C-3), 73.31 (C-4), 73.11 (C-2), 72.60
(C-30), 71.85 (C-20), 70.13 (C-50), 69.93 (C-5), 66.41 (C-6), 60.79
(C-60).
4.2.10. CGTase catalyzed glycosylations
0
0
0
0
Reactions catalyzed by B. macerans CGTase were performed in
sodium phosphate buffer (50 M, pH 7.0) at 40 °C. Thermoanaerob-
acter sp. CGTase-catalyzed reactions were carried out in BisTris–
HCl buffer (50 mM, pH 6.0) at 50 °C. If not specified otherwise,
20 mM b-cyclodextrin and 100 mM of p-nitrophenyl-b-
ranoside were used as substrate and 2.5 U mlꢀ1 of the respective
enzyme. Reactions with the -anomeric glucopyranoside as glyco-
syl-acceptor were carried out applying 10 mM b-cyclodextrin and
D-glucopy-
a
4.2.7. UV spectroscopic analysis of liberation of p-nitrophenol
The liberation of p-nitrophenol may occur as a slow non-cata-
lyzed side reaction or catalyzed by CGTase via hydrolysis or the
acceptor reaction. p-Nitrophenol concentrations were measured
immediately after appropriate dilution of samples in sodium phos-
phate buffer (50 mM, pH 7.0) via UV spectroscopy at 410 nm. Cal-
ibration using p-nitrophenol standard solution was performed in
the same buffer.
50 mM p-nitrophenyl-a-D-glucopyranoside due to the lower solu-
bility of the acceptor and 1 U mlꢀ1 of the respective CGTase. Sam-
ples were taken at given intervals and the reactions stopped by
quenching with 0.2 M HCl.
4.2.11. Purification of p-nitrophenyl-disaccharide isomers
Separation of p-nitrophenyl-maltooligosaccharides with differ-
ent number of glucose-units could be achieved applying a Reveleris
Amino-column (12 g) with gradient elution using acetonitrile:
water (50 mM acetic acid) at 18 ml minꢀ1 (Fig. S6, Supporting
4.2.8. Determination of CGTase activity and stability
CGTase activity was determined using a UV spectroscopic assay
based on the detection of b-cyclodextrin/phenolphthalein complex
formation as previously described by Higuti and coworkers on a
Beckman DU 650 spectrophotometer (Beckman Coulter, Brea, CA,
USA).39 The enzymatic reaction was started by addition of 0.1 ml
of properly diluted B. macerans CGTase to 1.4 ml of sodium phos-
phate buffer (50 M, pH 7.0) containing 1% soluble starch at 40 °C.
The same assay using BisTris–HCl buffer (50 mM, pH 6.0) at 50 °C
was applied for Thermoanaerobacter sp. CGTase. Reactions were
stopped by adding 0.05 ml of 0.1 M NaOH to 0.15 ml of the reaction
solution. 1 ml of a solution containing freshly prepared 0.06 mM
phenolphthalein in NaHCO3/Na2CO3 buffer (125 mM, pH 10.5)
was subsequently added, the solution mixed by vortexing and
measured immediately at 550 nm. Enzyme stability was deter-
mined by incubation of B. macerans CGTase in sodium phosphate
buffer, 50 M, pH 7.0 at 40 °C or Thermoanaerobacter sp. CGTase in
BisTris–HCl buffer, 50 mM, pH 6.0 at 50 °C, respectively, and subse-
quent measurement of remaining activity using the method
described above.
information). No separation of p-nitrophenyl-b-
D-nigeroside
(1?3-isomer) and p-nitrophenyl-b- -maltoside (1?4) could be
D
achieved using this method. Therefore, the reaction products as
obtained from CGTase catalyzed glycosylations were subjected to
the glucoamylase-catalyzed selective hydrolysis of
oligosaccharides (Fig. S5, Supporting information). Remaining
p-nitrophenyl-b- -nigeroside and p-nitrophenyl-b- -isomaltoside
a(1?4)-linked
D
D
were separated applying the same method as described above
using an Amino-column with gradient elution (Fig. S7, Supporting
information).
Acknowledgements
Financial support was provided by CONACYT (Grant No.
00154194 to G.S.R.) by PAPPIIT (Grant IN220208 to G.S.R.). NMR
spectroscopic data were collected with kind assistance of M.C. Sil-
via Marquina-Bahena from CIQ-UAEM. The authors also thank
Rodrigo Arreola-Barroso for his assistance with structural figures
preparation and Fernando González from IBT-UNAM for technical
assistance.
4.2.9. Enzyme purification
CGTases from B. macerans and Thermoanaerobacter sp. were
purified via affinity chromatography using
a-cyclodextrin modi-
Supplementary data
fied sepharose support. -Cyclodextrin was covalently coupled to
a
Sepharose 6B (Sigma–Aldrich, St. Louis, MO, USA) following the
protocol described by Sundberg and Porath40 and packed into a
10 ml column. A Reveleris X2 Flash Chromatography System
(Grace, Deerfield, IL, USA) was used for protein purification. Ini-
tially, the column was equilibrated with 10 column volumes (CV)
of sodium acetate buffer 10 mM, pH 5.5 at 2 ml minꢀ1. After man-
ual application of the CGTase sample (10 mg crude protein), the
column was washed with 10 CV of sodium acetate buffer 10 mM,
pH 5.5 and the CGTase eluted with the same buffer containing
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.carres.2014.11.
010.
References
1. van der Veen, B. A.; van Alebeek, G.-J. W. M.; Uitdehaag, J. C. M.; Dijkstra, B. W.;
Dijkhuizen, L. Eur. J. Biochem. 2000, 267, 658–665.
2. Martin, M. T.; Alcalde, M.; Plou, F. J.; Dijkhuizen, L.; Ballesteros, A. Biotechnol.
Bioeng. 2000, 19, 21–35.
3. Kitahata, S.; Hara, K.; Fujita, K.; Nakano, H.; Kuwahara, N.; Koizumi, K. Biosci.,
Biotechnol., Biochem. 1992, 56, 1386–1391.
4. Kitahata, S.; Okada, S.; Fukui, T. Agric. Biol. Chem. 1978, 42, 2369–2374.
5. Svensson, D.; Ulvenlund, S.; Adlercreutz, P. Green Chem. 2009, 11, 1222–1226.
6. Do, H.; Sato, T.; Kirimura, K.; Kino, K.; Usami, S. J. Biosci. Bioeng. 2002, 94, 119–
123.
7. Sato, M.; Matsuo, T.; Orita, N.; Yagi, Y. Biotechnol. Lett. 1991, 13, 69–74.
8. Nakano, H.; Kitahata, S.; Tominaga, Y.; Kiku, Y.; Ando, K.; Kawashima, Y.;
Takenishi, S. J. Ferment. Bioeng. 1992, 73, 237–238.
9. Torres, P.; Poveda, A.; Jimenez-Barbero, J.; Luis Parra, J.; Comelles, F.;
Ballesteros, A. O.; Plou, F. J. Adv. Synth. Catal. 2011, 353, 1077–1086.
10. Leemhuis, H.; Kelly, R. M.; Dijkhuizen, L. Appl. Microbiol. Biotechnol. 2010, 85,
823–835.
10 mg mlꢀ1
pooled and ultrafiltrated using an Amicon Ultra-15 Milipore
MWCO 10.000 spincolumn for buffer change and removal of
a-cyclodextrin. Fractions containing CGTase were
a
-
cyclodextrin. Enzymes were stored at 4 °C in BisTris buffer
(50 mM, pH 6.0) in the case of Thermoanaerobacter sp. CGTase or
sodium phosphate buffer (50 mM, pH 7.0) in the case of B. macer-
ans CGTase. Thermoanaerobacter sp. CGTase could not be purified to
homogeneity as visualized by SDS–PAGE (Fig. S8a, Supporting
information). However, the partially purified preparation of Ther-
moanaerobacter sp. CGTase showed the same reaction profile as
the pure enzyme isolated by gel extraction from the respective
polyacrylamide gel. It is assumed that the observed fragments in
11. Kim, Y.-H.; Lee, Y.-G.; Choi, K.-J.; Uchida, K.; Suzuki, Y. Biosci., Biotechnol.,
Biochem. 2001, 65, 875–883.

54
S. Strompen et al. / Carbohydrate Research 404 (2015) 46–54
12. Aga, H.; Yoneyama, M.; Sakai, S.; Yamamoto, I. Agric. Biol. Chem. 1991, 55,
1751–1756.
29. Knegtel, R. M. A.; Strokopytov, B.; Penninga, D.; Faber, O. G.; Rozeboom, H. J.;
Kalk, K. H.; Dijkhuizen, L.; Dijkstra, B. W. J. Biol. Chem. 1995, 270, 29256–29264.
30. Knegtel, R. M. A.; Wind, R. D.; Rozeboom, H. J.; Kalk, K. H.; Buitelaar, R. M.;
Lubbert Dijkhuizen, L.; Dijkstra, B. W. J. Mol. Biol. 1996, 256, 611–622.
31. Kubota, M.; Matsuura, Y.; Sakai, S.; Katsube, Y. Denpun Kagaku 1991, 38, 141–
146.
32. Lawson, C. L.; van Montfort, R.; Strokopytov, B.; Rozeboom, H. J.; Kalk, K. H.; de
Vries, G. E.; Penninga, D.; Dijkhuizen, L.; Dijkstra, B. W. J. Mol. Biol. 1994, 236,
590–600.
33. Leemhuis, H.; Rozeboom, H. J.; Wilbrink, M.; Euverink, G.-J. W.; Dijkstra, B. W.;
Lubbert Dijkhuizen, L. Biochemistry 2003, 42, 7518–7526.
34. Schmidt, A. K.; Cottaz, S.; Driguez, H.; Schulz, G. E. Biochemistry 1998, 37, 5909–
5915.
13. Kuriki, T.; Imanaka, T. J. Biosci. Bioeng. 1999, 87, 557–565.
14. Kobayashi, S.; Shibuya, N. Japan Kokai Tokkio Koho, JP63-216492, 1987.
15. Moreno, A.; Damian-Almazo, J. Y.; Miranda, A.; Saab-Rincon, G.; Gonzalez, F.;
Lopez-Munguia, A. Enzyme Microb. Technol. 2010, 46, 331–337.
16. Urbach, C.; Halila, S.; Guerreiro, C.; Driquez, H.; Mulard, L. A.; Armand, S.
ChemBioChem 2014, 15, 293–300.
17. Kometani, T.; Nishimura, T.; Nakae, T.; Takii, H.; Okada, S. Biosci., Biotechnol.,
Biochem. 1996, 60, 645–649.
18. Suzuki, Y.; Sizuki, K. Agric. Biol. Chem. 1991, 55, 181–187.
19. Kometani, T.; Terada, Y.; Nishimura, T.; Takii, H.; Okada, S. Biosci., Biotechnol.,
Biochem. 1994, 58, 1990–1994.
20. Li, D.; Roh, S.-A.; Shim, J.-H.; Mikami, B.; Baik, M.-Y.; Park, C.-S.; Park, K.-H. J.
Agric. Food Chem. 2005, 53, 6516–6524.
21. Crout, D. H. G.; Howarth, O. W.; Singh, S.; Swoboda, B. E. P.; Critchley, P.;
Gibson, W. T. J. Chem. Soc., Chem. Commun. 1991, 1550.
22. Gillard, B. K.; Markman, H. C.; Feig, S. Clin. Chem. 1977, 23, 2279–2282.
23. Mackenzie, L. F.; Wang, Q.; Warren, R. A. J.; Withers, S. G. J. Am. Chem. Soc. 1998,
120, 5583–5584.
24. Dupuy, G.; Hillaire, G.; Aubry, C. Clin. Chem. 1987, 33, 524–528.
25. Yoon, S.-H.; Robyt, J. F. Carbohydr. Res. 2006, 341, 210–217.
26. Ishii, N.; Haga, K.; Yamane, K.; Harata, K. J. Mol. Recognit. 2000, 13, 35–43.
27. Kanai, R.; Haga, K.; Yamane, K.; Harata, K. J. Biochem. 2001, 129, 593–598.
28. Klein, C.; Hollender, J.; Bender, H.; Schulz, G. E. Biochemistry 1992, 31, 8740–
8746.
35. Strokopytov, S.; Knegtel, R. M. A.; Penninga, D.; Rozeboom, H. J.; Kalk, K. H.;
Lubbert Dijkhuizen, L.; Dijkstra, B. W. Biochemistry 1996, 35, 4241–4249.
36. Uitdehaag, J. C.; Mosi, R.; Kalk, K. H.; van der Veen, B. A.; Dijkhuizen, L.;
Withers, S. G.; Dijkstra, B. W. Nat. Struct. Biol. 1999, 6, 432–436.
37. Uitdehaag, J. C.; Kalk, K. H.; van der Veen, B. A.; Dijkhuizen, L.; Dijkstra, B. W. J.
Biol. Chem. 1999, 274, 34868–34876.
38. Wind, R. D.; Uitdehaag, J. C. M.; Buitelaar, R. M.; Dijkstra, B. W.; Dijkhuizen, L. J.
Biol. Chem. 1998, 273, 5771–5779.
39. Higuti, I. H.; Anunciação da Silva, P.; Papp, J.; Mayumi de Eiróz Okiyama, V.;
Alves de Andrade, E.; Abreu-Marcondes, A. de.; Nascimento, A. J. do. Braz. Arch.
Biol. Technol. 2004, 47, 837–841.
40. Sundberg, L.; Porath, J. J. Chromatogr. 1974, 90, 87–98.