The method of integrated kinetics and its applicability to the exo-glycosidase-catalyzed hydrolyses of p-nitrophenyl glycosides

Article abstract of DOI:10.1016/j.carres.2015.03.021

In the present work we suggest an efficient method, using the whole time course of the reaction, whereby parameters k_cat, K_m and product K_I for the hydrolysis of a p-nitrophenyl glycoside by an exo-acting glycoside hydrola

Full text of DOI:10.1016/j.carres.2015.03.021

Carbohydrate Research 412 (2015) 43e49
Contents lists available at ScienceDirect
Carbohydrate Research
journal homepage: www.elsevier.com/locate/carres
The method of integrated kinetics and its applicability to the
exo-glycosidase-catalyzed hydrolyses of p-nitrophenyl glycosides
,
Anna S. Borisova a b, Sumitha K. Reddy c, Dina R. Ivanen a, Kirill S. Bobrov a,
,
Elena V. Eneyskaya a, Georgy N. Rychkov a e, Mats Sandgren b, Henrik Stålbrand c,
,
e
,
, e
Michael L. Sinnott d, Anna A. Kulminskaya a *, Konstantin A. Shabalin a
a National Research Center “Kurchatov Institute”, B.P. Konstantinov Petersburg Nuclear Physics Institute, 188300, Gatchina, Orlova roscha, Russia
b Department of Chemistry and Biotechnology, Swedish University of Agricultural Sciences, Uppsala, Sweden
c Department of Biochemistry and Structural Biology, Lund University, Lund S-221 00, Sweden
d Department of Chemical Sciences, University of Huddersfield, Queensgate, Huddersfield HD1 3DH, UK
e St. Petersburg State Polytechnical University, 29 Politekhnicheskaya st., 195251 St. Petersburg, Russia
a r t i c l e i n f o
a b s t r a c t
Article history:
In the present work we suggest an efficient method, using the whole time course of the reaction,
whereby parameters kcat, Km and product KI for the hydrolysis of a p-nitrophenyl glycoside by an exo-
acting glycoside hydrolase can be estimated in a single experiment. Its applicability was demonstrated
Received 28 November 2014
Received in revised form
23 March 2015
Accepted 27 March 2015
Available online 1 May 2015
for three retaining exo-glycoside hydrolases,
from Penicillium sp. and -galactosidase from Thermotoga maritima (TmGalA). During the analysis of the
reaction course catalyzed by the TmGalA enzyme we had observed that a non-enzymatic process,
mutarotation of the liberated -galactose, affected the reaction significantly.
© 2015 Elsevier Ltd. All rights reserved.
b-xylosidase from Aspergillus awamori, b-galactosidase
a
a-D
Keywords:
Integrated kinetics
Retaining glycoside hydrolase
Mutarotation
1. Introduction
depletion in the first instance. Nevertheless, quite often accumu-
lation of the product(s) and depletion of the substrate(s) cannot be
neglected.
By ‘integrated kinetics’ we mean use of the MichaeliseMenten
equation in its integrated form. First described by Schwert,1 its
successful application to the lactate dehydrogenase-catalyzed
reduction of pyruvate was published in 1969.2 Despite the theory
being summarized in a classic text,3 it has not become widely used,
and most enzymologists prefer to use initial velocities for esti-
mating kinetic parameters. This method, described by Michaelis a
century ago4 is convenient because it allows product accumulation
and substrate consumption to be ignored. Indeed, in the initial
stages of enzyme kinetic studies, the main task is to understand the
form of the kinetic law, and to locate the ranges within which ki-
netic parameters lie, ignoring product accumulation and substrate
Traditionally, in glycosidase assays, model substrates are used,
in which the glycone is a monosaccharide and the aglycone is a
chromogenic or a fluorogenic group. Kinetic constants can then be
easily determined from initial velocities measured by absorbance
or fluorescence. Where estimation of initial rates is difficult (e.g.,
because a few percent reaction at substrate concentrations at or
below the Km gives very small absorbance changes), kinetic pa-
rameters can be more reliably derived from analysis of a whole
kinetic curve. At the same time, the method of integrated kinetics
enables determination of all the constants needed to characterize a
reaction with a relatively small set of experiments. Using the in-
tegral form of the MichaeliseMenten equation, one can determine
inhibition type, the corresponding parameters Km, KI, kcat by vary-
ing initial substrate concentrations.
Abbreviations: TmGalA,
nitrophenyl; pNP Gal, pNP-
pyranoside; pNP Gal, p-nitrophenyl
a
a
-galactosidase from Thermotoga maritima; pNP, para-
-galactopyranoside; pNPXyl, p-nitrophenyl -xylo-
-galacopyranoside.
a
b
-
D
b
Here, we suggest a method of integrated kinetics to estimate
parameters kcat, Km and KI for the hydrolysis of a p-nitrophenyl
glycoside by an exo-acting glycoside hydrolase in a single experi-
b
* Corresponding author. National Research Center “Kurchatov Institute”, B.P.
Konstantinov Petersburg Nuclear Physics Institute, 188300, Gatchina, Orlova roscha,
Russia. Tel./fax: þ7 8137132014.
ment, and demonstrate its applicability for three enzymes,
b-
xylosidase from Aspergillus awamori (GH3; (CAZy; http://www.
E-mail address: kulm@omrb.pnpi.spb.ru (A.A. Kulminskaya).
http://dx.doi.org/10.1016/j.carres.2015.03.021
0008-6215/© 2015 Elsevier Ltd. All rights reserved.

44
A.S. Borisova et al. / Carbohydrate Research 412 (2015) 43e49
cazy.org),5,6
b
-galactosidase from Penicillium sp. (GH35)7 and
a
-
In case of the inhibition by a product, the concentration of which
is changing during the reaction course, the expression for the re-
action velocity can be written using equation of a material balance
S0¼SþP and assuming that S0>>E0:
galactosidase from Thermotoga maritima (GH36).8,9 During the
analysis of the reaction course catalyzed by the
from Thermotoga maritima (TmGalA) we had observed that a non-
enzymatic process, mutarotation of the liberated -galactose,
affected the derived parameters. This effect was reported previ-
ously for the
-galactosidase-catalyzed hydrolysis of lactose10e12
but, to our knowledge, never for -galactosidases. In the present
work, a kinetic model invoking simultaneous competitive inhibi-
tion of the exo-glycoside hydrolase by the reaction product and by
its spontaneously-formed anomers, with different KI values, is
suggested for the first time.
a-galactosidase
a-D
kcatE0S
ꢀ
ꢁ
VIP
¼
(4)
b
S0ꢀS
S þ Km 1 þ
KI
a
After simple transformations the Eq. 4 may be rewritten as
app
k
E0S
cat
VIP
¼
(5)
(5a)
(5b)
S þ Kmapp
2. Results and discussion
where
Km
KI
Km
1 ꢀ
KI
One of the possible mechanisms for enzymatic hydrolysis can be
described with Scheme 1, when competitive inhibition takes place.
The reaction includes formation of the Michaelis complex ES
followed by release of the product P and regeneration of the
enzyme E. In this case the hydrolysis rate is described by the well-
known MichaeliseMenten equation:4
Km þ S0
Kmapp
¼
and
kcat
app
cat
k
¼
:
Km
KI
1 ꢀ
kcatE0S
Km þ S
V ¼
(1)
The equations 5aand 5b are valid for the case KmsKI, otherwise
a dominator is zero and the whole expression becomes undeter-
mined. However, in the case Km¼KI, the velocity V can be expressed
as KꢁS where K is the first order rate constant equal kcat/(KmþS0).
One can easily distinguish this particular case from the first order
reaction due to dependence of K on the initial substrate concen-
tration S0. Eq. 3 is well known3 though, as a rule, it has been
expressed in another form. In general, at a single initial substrate
concentration, Eq. 3 makes it impossible to distinguish the effect of
product inhibition from any other retarding process, whereas
varying initial concentrations of the substrate enables such
discrimination (a review of analytic methods of time-course of
enzyme reactions are given by Cornish-Bowden3). Here, we suggest
a convenient and clear method to determine kinetic parameters
based on Eq. 5a. Indeed, integration of Eq. 5 leads to Eq. 3; values for
where V is a reaction velocity, S, E0 are corresponding initial con-
centrations of a substrate and enzyme, kcat, Km are kinetic param-
eters of the reaction: the graph describes a rectangular hyperbola
The MichaeliseMenten Eq. 1 can be integrated1 and written as Eq.
2:
ꢀ
ꢁ
S0
S
kcatE0t ¼ S ꢀ S0 þ Km ln
(2)
The Eqs. 1 and 2 are valid under steady-state conditions, when
the flux between all forms of the enzyme is much less than the flux
between the product and the substrate. For the method we used in
the present work one needs to consider two issues. First, in case of a
competitive inhibition at fixed inhibitor concentrations, a form of
the Eqs. 1 or 2 remains unchanged while Km is increased by a factor
of (1þI/KI) and kcat is not altered.3 So, Eq. 2 can be rewritten as:
Kmapp and kapp may be easily obtained by non-linear fitting of the
cat
whole kinetic curve. Finally, from the dependence of Kmapp on the
initial concentration of the substrate S0 (Eq. 5a) one can derive two
parameters, Km and KI.
ꢀ
ꢁ
S0
S
app
cat
k
E0t ¼ S ꢀ S0 þ Kmapp ln
(3)
It is clear that Kmapp is a linear function of the initial substrate
where Kmapp and kapp are apparent kinetic parameters that relate to
true Km and kcat by equations
concentration S0 with
a
slope Km=KI ꢀ Km and y-intercept
cat
app
KmappðS0Þ ¼ Km=1 ꢀ . Therefore, the linear dependence of Km on
Km
KI
So provides a convenient method of analysis. A set of time courses
at various values of So and their fitting to Eq. 3 will result in a set of
app
cat
Kmapp ¼ Km ꢁ ð1 þ I=KIÞ;
k
¼ kcat
:
Kmapp and kapp values. Linear regression of Kmapp as a function of So
From this, analyzing dependency Kmapp on inhibitor concentra-
tion, one can derive KI and Km, traditionally in form of a Dixon
plot.13
cat
yields two parameters, the slope A and the y-intercept B. Using Eq.
5a one can derive kinetic constants Km ¼ B=1 þ A and KI ¼ B and
the catalytic constant kcat ¼ kcat
approach enables one to determine kcat, Km and KI in a simple
experiment with variation of S0.
=
A
appꢂ
from Eq. 5b. So, this
1 þ A
To test the methodology, we used enzymes available in our
laboratory, the
b
-xylosidase from A. awamori,5,6 the
b
-galactosidase
from Penicillium sp.7 and the
a
-galactosidase from Thermotoga
maritima.8,9 Our aim was to analyze a dependence KmappðS0Þ for
every enzyme and compare values for kinetic parameters obtained
by two methods: the widely used analysis of the initial velocities
and the method of the integral kinetics (eq. 5a). All three enzymes
produce the aldopyranose product in the same anomeric configu-
ration as the substrate. All enzymes probably act as exo-glycosi-
dases with only one glycone-binding subsite (i.e., the ꢀ1
Scheme 1. Minimal kinetic scheme for hydrolysis reaction with competitive inhibition,
following MichaeliseMenten mechanism.

A.S. Borisova et al. / Carbohydrate Research 412 (2015) 43e49
45
subsite)6,7,9 and the monosaccharide product of the hydrolysis of
pNP glycoside by each of these enzymes may very well be a
competitive inhibitor for the corresponding reaction.
For the b-D-xylosidase and b-D-galactosidase kinetic properties
obtained by the method of initial velocities revealed xylose and
galactose were competitive inhibitors. Kinetic parameters obtained
by the two methods were generally comparable (Table 1). But what
was more important, dependences KmappðS0Þ were linear in both
cases and showed good reproducibility, with an error bar of less
than 2% (an example of such linear dependence for the b-galacto-
sidase is given in Fig. 1). At this stage the method seemed generally
successful.
In the case of TmGalA, we expected to see a similar linearity for
the KmappðS0Þ dependence. However, we observed variations in the
obtained values for Kmapp in the range of high initial substrate con-
centrations (5 mM and higher) at different enzyme concentrations
(Fig. 2): a threefold variation in TmGalA concentrations resulted in
differences between KI values obtained from Eq. 5a of 40%. Besides
the approaches used for the analysis of the data, the only difference
for each experiment was the length of time period, for which the
reaction was run. Our assumption was that the disagreement be-
tween KI values may be caused by another process accompanying
Fig. 1. Kampp dependence on S0 concentration in pNP
galactosidase from Penicillium sp.
bGal hydrolysis catalyzed by the b-
the enzymatic reaction. We supposed that the mutarotation of a-D-
galactopyranose released by the enzyme during the hydrolytic re-
action affects the kinetic parameters in the reaction course, and
both
a
- and
b
-galactopyranose, and traces of an acyclic form, two
- and -galactofuranose, are involved in the
anomeric furanoses,
a
b
mutarotation. This process is catalyzed by general acids and bases.16
Accurate modeling of galactose mutarotation is quite complicated,
therefore several assumptions were made to simplify the kinetic
model. First, we assumed furanose forms can be neglected due to
their presence in the amount of 4.9% (determined by NMR under
conditions of the kinetic experiments) of the total amount of
galactose and therefore the mixture consists generally of gal-
that the new forms generated by mutarotation (largely the
ranose form) had KI values different from -galactopyranose. A
similar effect was reported for the -galactosidase from Aspergillus
niger that distinguished between - and -galactose in the hydro-
lysis of lactose.10 Analysis of the X-ray structures of
-galactosidases
belonging to different GH families suggested different affinities of
these enzymes to - and -forms of -galactopyranose. Thus, the
crystal structure of the -galactosidase from Lactobacillus aci-
dophilus (36 GH14) revealed
-galactopyranose in the active
center, whereas the crystal structure of the 27 GH family -galac-
tosidase from Trichoderma reesei contained
-galactopyranose.15
To illustrate binding variants of - and -galactopyranose in the
b-py-
a-D
b
a
b
a
a
b
D
actopyranoses. Since the mixture of
actopyranose at the equilibrium consists approximately of 33.3%
form and 61.8%
-form at 37 ꢂC, these values were used for calcu-
a- and b-anomers of gal-
a
a
-
a-D
b
a
lations. Our data generally agree with those previously published,17
which pertain to slightly different conditions (100% of D2O at 31 ꢂC).
Second assumption was that furanoses would not influence
TmGalA kinetics. The ring closure to five membered rings is faster
than the closure to six-membered rings, so transient accumulation
of furanose forms could affect results, especially if the furanose
b-D
a
b
active center of TmGalA, they were assessed with flexible molecular
docking. Four different protonation states of the pair of catalytic
residues Asp327 (nucleophile) and Asp387 (acid/base) were
examined including: Asp327/Asp387, Ash327/Asp387, Asp327/
Ash387 and Ash327/Ash387 (Asp is charged aspartic acid, Ash is
neutral protonated aspartic acid). Binding propensities of the two
galactopyranose anomers were evaluated from calculating binding
scores. Results of the calculations are given in Supplemental data.
The interconversion of the a- and b-anomeric forms of D-gal-
actopyranose occurs as the sugar ring opens to the aldehydo form
and recloses, yielding the opposite anomeric configuration. Besides
Table 1
Kinetic parameters for the hydrolysis of the pNP glycoside by b-D-xylosidase and b-
D
-galactosidase
Kinetic parameters
Method of initial
velocities
Method of integrated
kinetics
b-D
-xylosidasea
Km, mM
kcat, sꢀ1
KI, mM
0.28 0.03
0.29 0.015
71 3
0.95 0.06
65
4
1.06 0.09
b-D
-galactosidaseb
Km, mM
kcat, sꢀ1
KI, mM
0.30 0.01
0.37 0.04
54
2
40
2
0.91 0.03
1.25 0.07
a
Experiments were carried out at pH 4.5 and 37 ꢂC, see details in sections 3.2 and
3.3 of Experimental.
b
Experiments were carried out at pH 5.0 and 37 ꢂC, see details in sections 3.2 and
Fig. 2. Kampp dependences on S0 concentration in pNP
TmGalA: - is the enzyme concentration E0¼0.25 M; B is E0¼0.75
a
Gal hydrolysis catalyzed by
M.
3.3 of Experimental.
m
m

46
A.S. Borisova et al. / Carbohydrate Research 412 (2015) 43e49
of furanose forms to a-D-pyranose decreases twofold, from 0.148 to
0.073 However, at 4 ꢂC KI is approx. the same as at 37 ꢂC and thus,
binding with furanose does not, probably, affect the inhibition due
to noticeable decrease of concentration of furanose forms.
To demonstrate the effect of spontaneous mutarotation of the
product on the time-courses of TmGalA-hydrolysis of pNPaGal, we
altered the enzyme concentration so that a high conversion of
substrate was attained over widely differing time intervals. When
the reaction time did not exceed 5 min (achieved by the use of
stopped flow equipment), the time-courses were superimposable
when the time axis was normalized by the enzyme concentration
(i.e., one set of curves with an enzyme concentration Eo and tobs as
the x axis, another set of curves with 2ꢁEo and 2ꢁtobs as the x axis)
by time proportional to enzyme concentration (Fig. 3). It means
that mutarotation does not affect the total reaction rate and only
enzymatic hydrolysis and any inhibition by
a-galactopyranose
takes place. Otherwise, under long-term reaction conditions (more
than 15 min) normalized enzyme kinetic curves were not imposed
(Fig. 4B).
The experiment with various concentrations of the enzyme
enabled us to assess the inhibitory power of
anomers. At the highest TmGalA concentration, the flux of the
enzymatic reaction far exceeded the flux from - to -galactose
from spontaneous reaction, and -galactose remained predomi-
nant in the reaction mixture. Conversely, the equilibrated mixture
of - and -anomers of galactose (33.3 and 61.8%, respectively)
a- or b-galactose
Fig. 3. Normalized enzyme kinetic curves for pNPaGal hydrolysis by TmGalA. Initial
substrate concentration was 1 mM; (each time point is multiplied by ratio of enzyme
concentrations ki¼Ei/E1): B is E1¼0.17 M, - is E2¼0.08 M, V is E3¼0.04 M.
a
b
m
m
m
a
a
b
forms bound more tightly than pyranose forms. However, the
equilibrium between -pyranose and all furanose forms is
reached faster than the equilibrium between - and
-pyranoses.18
Therefore, the ratio of -pyranose to furanoses is kept unchanged
during the whole process and binding of the TmGalA with any of
the furanose forms is proportional to -galactopyranose con-
accumulated at the smallest enzyme concentration when the
enzymic flux was less than the mutarotation flux. Therefore if one
superimposes such normalized curves, the curves diverge. In the
first case, the reaction rate did not depend on either the enzyme or
a-D
a
b
a-D
a
- and
slower reaction in normalized coordinates corresponds to stronger
inhibition and it would mean the higher affinity of -anomer
b-galactose concentration in contrast to the second case. The
a-D
centration and is indistinguishable in a kinetic experiment. Mea-
surement of the equilibrium of tautomers under the kinetic
a
(lower value for KI). Fig. 5B shows this: apparent acceleration at
around 50% reaction can only be brought about in an enzyme we
have shown to be Michaelian by spontaneous removal of an in-
experiments (50 m
М NaAc, pH 5.0) showed that at temperature
lowering, the equilibrium between tautomeric forms is changing:
at 25 ꢂC the amount of
was 2.84%; at 4ꢂС:
p was 26.04%,
a
p was 33%,
b
b
p was 62.14%,
p was 72.0%,
a
f was 2.04%,
b
b
f
f
hibitor to a less inhibitory species, in this case
Therefore we can conclude that in cases of the fastest and slowest
reactions inhibition occurs by -anomer and by the equilibrated
b-galactopyranose.
a
af was 0.81%,
was 1.1%. As the temperature is lowered from 25 ꢂC to 4 ꢂC the ratio
a
Fig. 4. Effect of D-galactose interconversion in the hydrolysys of pNPaGal by TmGalA. A: Enzyme kinetics for pNPaGal hydrolysis by TmGalA with fixed initial substrate con-
centration at 1 mM, reaction temperature was 37 ꢂC. B: Normalized enzyme kinetic curves from plot A (each time point is multiplied by ratio of enzyme concentrations ki¼Ei/E1). For
both plots B: E1¼0.55
mM, -: E2¼0.28
mM V: E3¼0.14
mM, C: E4¼0.07
mM, ,: E5¼0.03
mM.

A.S. Borisova et al. / Carbohydrate Research 412 (2015) 43e49
47
I
Ia
Ib
KIb
¼
þ
;
KI KIa
!
(6)
Ia
KIa
Ib
KIb
Kmapp ¼ Km 1 þ
þ
Ia$KIa$KI
KIa ꢀ Ia$KI
KIb
¼
(7)
Ia ¼ 0:265I
Ib ¼ 0:735I
Finally, we compared kinetic parameters obtained previously9
with results of calculations derived from the integrated rate Eqs.
5a and 5b, considering galactose interconversion. The minimal ki-
netic scheme describing the pNPaGal hydrolysis by TmGalA is
presented in Scheme 2.
Fig. 5. Kinetics of D-galactose mutarotation monitored by NMR in situ at 37 ꢂC. For
-D-galactopyranoseeC, for -D-galactopyranoseeB.
a
b
mixture (
a
-þ
b-), respectively. When the reducing sugar product is
fully equilibrated, the rate of mutarotation does not influence the
overall reaction. However, one should include mutarotation into
the minimal kinetic scheme of hydrolysis when comparable rates of
the enzymatic reaction and mutarotation (under certain tempera-
ture and pH) take place.
Since
reaction and mutarotaion is pH- and temperature-dependent, we
studied the kinetics of -/ -galactose interconversions under
D-galactose is a competitive inhibitor of TmGalA hydrolytic
a
b
defined conditions by NMR spectroscopy, yielding mutarotation
rates: ka ¼0.002 sꢀ1 kb ¼0.001 sꢀ1. Kinetic data (Fig. 5) shows
4b
4a
that the balance of a- and b-galactose can be reached within 25 min
at 37 ꢂC. Lowering the temperature to 4 ꢂC led to the decrease of
mutarotation rates approximately 50 times (ka ¼0.000045 sꢀ1
Scheme 2.
4
b
kb ¼0.000016 sꢀ1). Thus, we considered the mutarotation to be
4a
negligible at 4 ꢂC for 30 min, since the
b-galactose percentage does
not reach 7.5% of total galactose. A similar temperature dependence
was observed for galactose mutarotation by Fleschel et al.10 Thus,
incubation at 4 ꢂC made it possible to distinguish between inhibi-
where S is pNP
p-nitrophenol.
a a-D b-D-galactose; P is
Gal; Pa is -galactose; Pb is
For the scheme of the hydrolysis without inhibition, two in-
dependent catalytic constants have to be determined, i.e., kcat and
Km for the pNPaGal hydrolysis (equation A). When competitive
inhibition is considered, equations A and B are valid for fast or
tion by
hydrolysis of pNP
nas fimi
-mannanase hydrolysis was studied similarly.19
Further experiments with two types of inhibitors were carried
out on ice. Inhibitors were: a). -galactopyranose obtained by fast
dissolution of crystalline -galactose in water immediately before
a- and
b
-forms of
D-galactose in case of TmGalA-catalyzed
a
Gal. Mannose mutarotation during Cellulomo-
b
slow reactions. For the fastest reactions inhibition is determined
by the
a-anomer, in the slowest ones the equilibrated mixture (a-
-) inhibits the reaction. When comparable rates of the enzy-
a-D
þb
a-D
matic reaction and mutarotation take place, equations C and D
should be added into the minimal kinetic scheme of the
hydrolysis.
Data from the Table 3 demonstrate that the KI value obtained
using the integrated form of Michaelis kinetics differs from those
determined by the standard Dixon method, because normally one
the experiment, and b). an equilibrated solution of the a- and b-D-
galactoses. The values for KIa and KIaþb were calculated and are
shown in the Table 2. Value for KIa in inhibition studies with
equilibrated mixture of
a- and b-anomers was calculated in the
assumption that the equilibrated concentration of a-D-galactose is
26.5% of a total initial one. Value for KIb was calculated according to
uses approximately equilibrated mixture of
inhibition studies.
a- and b-galactose in
the Eq. 7, derived from Eq. 6 valid for the inhibition experiment
with equilibrated mixture of a- and b-D-galactose.
In conclusion, the method of integrated kinetics used in the
present work can be easily applied for analysis of the glycoside
hydrolase-catalyzed hydrolysis of pNP glycosides. However, where
there is product inhibition, and mutarotation takes place on a
timescale comparable to the period for which the reactions are
followed, more complex treatments must be applied, particularly,
as with TmGalA, where the first-formed anomer binds more tightly
than the products of its mutarotation, yet these dominate the
mutarotational equilibrium. Thus, a more complicated reaction
model including mutarotation is required for determination of ki-
netic parameters of the hydrolysis.
Table 2
Inhibition constants for anomers of the
pNP Gal
D-galactopyranose in the hydrolysis of
a
Kampp, mM
KaI þb, mM
KaI , m
М
KbI , m
М
No inhibitor
-galactose
)-galactose
0.070 0.005
0.32 0.04
0.16 0.01
a
0.28 0.03
0.28 0.03
(
aþb
0.78 0.05
2.2 0.5
Reactions were carried out for 20 min, at 4 ꢂC.

48
A.S. Borisova et al. / Carbohydrate Research 412 (2015) 43e49
Table 3
Kinetic parameters of pNP
a
Gal hydrolysis obtained by methods of the initial velocities and integrated MichaeliseMenten kineticsa
Kinetic parameters
MichaeliseMenten-type kinetics
Method of the integrated kinetics
(not considering mutarotation)b
Method of the integrated kinetics
(considering mutarotation)
KM, mM
kcat, sꢀ1
KI, mM
0.080 0.005c
8.6 0.6c
0.060 0.002
9.1 0.3
0.63 0.02
0.060 0.002
9.1 0.3
0.80 0.06d
KaI ¼0.32 0.02 mM;
KbI ¼5.7 0.1 mM
a
Experiments were carried out at pH 5.0 and 37 ꢂC.
The reaction was carried out for 30 min that corresponded approx. 90% conversion of the substrate.
Data from the Ref. 9.
b
c
d
Data obtained by Dixon method.13
3. Experimental
in 50 mM sodium acetate buffer pH 5.0. The reaction catalyzed by
TmGalA was carried out in 50 mM sodium acetate buffer pH 5.0 at
37 ꢂC, the best compromise between maintaining activity of the
thermophile-derived enzyme and minimizing spontaneous hy-
3.1. Chemicals
All chemicals were obtained from Sigma Chemical (St. Louis,
USA) or Acros Organics (Geel, Belgium) unless otherwise noted, and
drolysis of pNP
aliquot of the enzyme (0.2e1.7 mU) to the corresponding substrate
solution (105e850 M, total reaction volume 320 L).
aGal. Hydrolysis was initiated by adding a 50 mL
were used without further purification. Substrates p-nitrophenyl
-galactopyranoside and p-nitrophenyl -xylopyranoside were
synthesized from the corresponding monosaccharides according to
the method described for -gal-
-glucosides;20 p-nitrophenyl
actopyranoside (pNP Gal) was synthesized from -galactose as
previously described for
-glucoside.21
b
-
m
m
D
b
-D
Kinetic curves for the hydrolysis of the corresponding pNP
glycoside by each enzyme were recorded with a Hitachi U-3310
spectrometer equipped with a circulating water bath, using 1 cm-
path length quartz cuvettes. Specific conditions for each
glycosidase-catalyzed reactions are given above. The accumulation
of p-nitrophenol during the reaction course for 20e50 min was
monitored at 365 nm relative to a reference sample containing no
enzyme. Each measurement was repeated more than three times.
Progress curves for enzyme kinetics were fitted using DYNAFIT
software22 with simple MichaeliseMenten mechanism without
inhibition as the input model. KaI pp (S0), kcaaptp(S0) plots were done
using the program Origin 8.0 (OriginLab Corp., Northampton, MA).
In the DYNAFIT package the symbolism of chemical equations
was translated into the underlying systems of ordinary differential
equations (ODE) by using the theory of matrices. Each kinetic curve
was described by the set of parameters: rate constants, analytic
concentrations of reactants, molar response coefficients. Analytic
concentrations and instrumental offset of reactants were set vari-
able for the better least-squares minimization. Molar absorption
coefficient for p-nitrophenol ε¼1.48ꢁ103 Mꢀ1ꢁcmꢀ1 (pH 5.0,
b
a-D
a
D
a
3.2. Enzymes and enzymes assays
The
fied as described previously.5,6 The
lium sp. was purified according to the procedure reported in Ref. 7 A
plasmid containing the -galactosidase gene was kindly donated
by Prof. R.M. Kelly (North Carolina State University, USA). The
b
-D-xylosidase from Aspergillus awamori X-100 was puri-
b-D
-galactosidase from Penicil-
a-D
a-D-
galactosidase gene from Thermotoga maritima MSB8 was expressed
in E. coli BL-21(DE3) and the recombinant enzyme was purified
from cells grown overnight as previously described.9
b
-Xylosidase activity towards p-nitrophenyl
(pNPXyl) was determined at 37 ꢂC in 50 mM sodium acetate buffer,
pH 4.5. One unit of the -xylosidase activity was defined as an
b-D-xylopyranoside
b-D
amount of the enzyme releasing 1
mM of p-nitrophenol from
l
¼365 nm) was used as a constant. The rate constants k1, k and
ꢀ1
pNPXyl per min.
kcat defined degree of freedom for the theoretical curve, therefore k1
was fixed (using value for a diffusion coefficient for small molecules
One unit of the
-galactopyranoside (pNP
enzyme required to hydrolyze 1
37 ꢂC in 50 mM sodium acetate buffer, pH 5.0.
The -galactosidase activity towards p-nitrophenyl
actopyranoside (pNP
Gal) was determined at 37 ꢂC in 50 mM so-
dium acetate buffer, pH 5.0. One unit of the activity was defined as
b
-D
-galactosidase activity towards p-nitrophenyl
Gal) was defined as the amount of the
M of the substrate per 1 min at
b-
D
b
in solution 103 mMꢀ1 ꢀ1) to provide better result of fitting.
s
m
Kinetic constants were calculated by solving the system of or-
dinary differential equations derived from the MichaeliseMenten
kinetics using steady-state assumptions with and without influ-
a-
D
a-D-gal-
a
ence of the
D-galactose inhibition. The model with mutarotation
included additional parameters ka
and kb as the constants,
4b
4a
amount of the enzyme releasing 1
mM of nitrophenol from pNPaGal
per min at 37 ꢂC.
which were obtained in an NMR experiment. The interconversion
of the - and -galactopyranoses was assumed to be the only
important reaction in the mutarotation: the - and -furanoses
a
b
MichaeliseMenten parameters in the hydrolysis of pNP glyco-
sides by -xylosidase, -galactosidase and -galactosidase
a
b
b-
D
b-
D
a-D
were observed to present to the extent of only 1.7 and 3.2%,
respectively.
were also determined according to the method of initial velocities
as described previously for each enzyme.6,7,9
The competitive inhibition constant for
D-galactose was deter-
mined from initial rates of hydrolysis of pNP
a
Gal obtained over
3.3. Enzyme kinetics
0.016e0.4 mM concentration and inhibitor concentration over
0.5e3 mM according to the Dixon method.13
Hydrolysis of pNP
37 ꢂC in 50 mM sodium acetate buffer pH 4.5. The process was
initiated by adding a 10 L aliquot of the enzyme (3.2 mU) to the
corresponding substrate solution (0.25e1.0 mM, total reaction
bXyl by the b-D-xylosidase was carried out at
m
3.4. Determination of D-galactose mutarotation rate
volume 200
sidase, a 50
m
m
L). For the hydrolysis of pNP
L aliquot of the enzyme (1.8e16.8 mU) was added to
L)
b
Gal by the
b
-
D
-galacto-
Mutarotation experiments were carried out using Varian NMR
700 MHz Spectrometer at 37 ꢂC, 25 ꢂC and 4 ꢂC in the buffer so-
lution routinely applied for kinetics (see above) with addition of
the substrate solution (0.3e1.3 mM, total reaction volume 320
m

A.S. Borisova et al. / Carbohydrate Research 412 (2015) 43e49
49
10% D2O. Crystalline
a
-
D
-galactopyranose was dissolved directly in
0.6e10 mU so that the complete hydrolysis time was not less than
15 min, and could be considered as ‘long-monitoring-time’ reac-
tion. Enzyme progress curves were fitted using DYNAFIT software.22
an NMR tube to the final concentration 1 mM, placed immediately
into the magnet followed by magnetic field homogeneity and sol-
vent signal suppression. To suppress the solvent (H2O) signals, the
pulse sequence with solvent presaturation from standard Varian
pulse sequence library was used (relaxation delay d1¼12.0 s,
acquisition time aq¼4.9 s, transition number nt¼4). 1H NMR
spectra were recorded each minute and concentrations of tauto-
mers were determined by integration of peaks corresponding to
3.7. Docking of
D
-galactose isomers in the active center of TmGalA
- and -galactose into the
Molecular modeling and docking of
a
b
active center of TmGalA were performed using the ICM Pro 3.7
program package.23,24 See details in the Supplemental data.
anomeric protons (chemical shifts (
for the - and -pyranose forms; 5.28 ppm and 5.22 ppm for the
and -furanose forms of -galactose, respectively). The curve plot
d) were 5.26 ppm and 4.58 ppm
a
b
a-
Acknowledgements
b
D
and calculations were made using Origin 8.0 software (OriginLab
Corporation, USA). To select relaxation delay d1, longitudinal
relaxation times T1 were measured using pulse sequence 'inver-
sion-recovery'. The longest relaxation time T1 was determined for
The work was supported by the Russian Foundation for Basic
Research (project No. 12-08-00813-a), a personal scholarship from
the Swedish Institute to A.B., and personal fellowship of the Gov-
ernment of Leningrad District, St. Petersburg, Russia to G.R. NMR
experiments were carried out using scientific equipment of the
Center of Shared Usage “The analytical center of nano- and bio-
technologies of Saint Petersburg State Polytechnical University”
financially supported by the Ministry of Education and Science of
the Russian Federation.
the
b
-furanose form at 37 ꢂC as 3.3 s and thus, recycling time
(aqþd1) was selected as 5T1 and equal 16.9 s.
3.5. Inhibition of TmGalA-catalyzed hydrolysis of pNPaGal by
anomers of D-galactose
Experiments were carried out with two types of inhibitor:
galactose and the fully equilibrated -galactose solution, at a final
concentration of 1 mM. To minimize mutarotation, -galactose
was dissolved in water kept on ice, and added immediately to the
reaction mixture. The equilibrated mutarotation mixture was ob-
a-D-
Supplementary data
D
a-D
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.carres.2015.03.021.
tained by incubation of freshly dissolved of
a
-
D
-galactose at room
Gal hydrolysis
temperature for 1 h and then cooled down. The pNP
a
References
reactions were carried out at 4 ꢂC in 20 mM sodium phosphate-
citrate buffer pH 5.0, with substrate concentrations between
0.031 mM and 2 mM. The reactions were initiated by addition of 10
1. Schwert GW. J Biol Chem 1969;244:1278e84.
2. Schwert GW. J Biol Chem 1969;244:1285e90.
3. Cornish-Bowden A. Fundamentals of Enzyme Kinetics. 4th ed. Weinheim, Ger-
many: Wiley-Blackwell; 2013.
mU of pre-cooled TmGalA, and stopped after 20 min by addition of
4. Johnson KA, Goody RS. Biochemistry 2011;50:8264e9.
5. Eneyskaya EV, Brumer 3rd H, Backinowsky LV, Ivanen DR, Kulminskaya AA,
Shabalin KA, et al. Carbohydr Res 2003;338:313e25.
1 mL of 10% Na2CO3; the subsequent protocol has been described
before.9 Inhibition constants were derived from the Line-
weavereBurk plots in the presence and absence of inhibitor.13
6. Eneyskaya EV, Ivanen DR, Bobrov KS, Isaeva-Ivanova LS, Shabalin KA,
Saveliev AN, et al. Arch Biochem Biophys 2007;457:225e34.
~
7. Neustroev KN, de Sousa EA, Golubev AM;, Brandao Neto JR, Eneyskaya EV,
3.6. Studies of influence of mutarotation on pNPaGal hydrolysis by
TmGalA
Kulminskaya AA, et al. Acta Cryst D 2000;D56:1508e9.
8. Miller ES, Parker KN, Liebl W, Lam D, Callen W, Snead MA, et al. Methods
Enzymol 2001;330:246e60.
9. Comfort DA, Bobrov KS, Ivanen DR, Shabalin KA, Harris JM, Kulminskaya AA,
et al. Biochemistry 2007;46:3319e30.
Rapid pNPaGal hydrolysis reactions were monitored by a Hi-
Tech Scientific KinetAsyst SF-61 Double Mixing Stopped Flow Sys-
tem (TgK Scientific), equipped with a circulating water bath. Re-
actions were carried out under routine conditions at fixed substrate
concentration at 1 mM. To minimize the effect of mutarotation,
hydrolysis was initiated by addition of TmGalA so that complete
reaction time did not exceed 5 min. Enzyme amounts in a range of
15e67 mU were used to obtain a set of integral kinetic curves
produced in five replicates. The accumulation of p-nitrophenol was
10. Flaschel E, Raetz E, Renken A. Biotechnol Bioeng 1982;24:2499e518.
11. Yang Sh-T, Okos MR. Biotechnol Bioeng 1989;34:763e73.
12. Peterson RS, Hill Jr CG, Amundson CH. Biotechnol Bioeng 1989;34:429e37.
13. Dixon M. Biochem J 1953;55:170e1.
14. Fredslund F, Hachem MA, Larsen RJ, Sørensen PG, Coutinho PM;, Lo Leggio L,
et al. J Mol Biol 2011;412:466e80.
~
15. Golubev AM, Nagem RA;, Brandao Neto JR, Neustroev KN, Eneyskaya EV,
Kulminskaya AA, et al. J Mol Biol 2004;339:413e22.
16. Capon B, Thacker D. J Chem Soc B 1967:1322e6.
17. Zhu Y, Zajicek J, Serianni AS. J Org Chem 2001;66:6244e51.
18. Kijima H, Go N. Carbohydr Res 1975;40:235e53.
recorded at 365 nm. Slow pNPaGal hydrolysis was monitored by an
€
19. Anderson L, Hagglund P, Stoll D;, Lo Leggio L, Drakenberg T, Stålbrand H. Biocat
Eon™ Microplate Spectrophotometer, with an absorbance micro-
plate reader (BioTek Instruments). Progress curves for enzyme ki-
netics were recorded at 365 nm and p-nitrophenol accumulation
was monitored as described in the Section 3.3 (Enzyme kinetics).
Reactions were carried out in a 96-well plate at a fixed substrate
concentration of 1 mM. The amount of enzyme was varied within
Biotrans 2008;26:86e95.
20. Dess D, Kleine HP, Weinberg DV, Kaufman RJ, Sidhu RS. Sintesis 1981;11:
883e5.
21. Helferich B, Schmitz-Hellebrecht E. Ber 1933;66:378e83.
22. Kuzmic P. Anal Biochem 1996;237:260e73.
23. Abagyan R, Totrov M. J Mol Biol 1994;235:983e1002.
24. Abagyan R, Totrov M, Kuznetsov D. J Comput Chem 1994;15:488e506.