Académique Documents
Professionnel Documents
Culture Documents
c o m p u t e r m e t h o d s a n d p r o g r a m s i n b i o m e d i c i n e x x x ( 2 0 1 5 ) xxxxxx
a r t i c l e
i n f o
a b s t r a c t
Article history:
2 December 2015
less, kinetic studies usually involved the evaluation of other inhibitors beyond the reaction
product. The occurrence of these situations emphasizes the importance of extending the
Keywords:
one inhibitor because reaction product is always present. This methodology is illustrated
with the reaction catalyzed by alkaline phosphatase inhibited by phosphate (reaction prod-
equations
uct, inhibitor 1) and urea (inhibitor 2). The approach is explained in a step by step manner
Linear inhibition
regression was performed with the Solver add-in (Microsoft Ofce Excel). Discrimination
of the kinetic models was carried out based on Akaike information criterion. This work
Two inhibitors
1.
Introduction
http://dx.doi.org/10.1016/j.cmpb.2015.12.013
0169-2607/ 2016 Elsevier Ireland Ltd. All rights reserved.
Please cite this article in press as: R.M.F. Bezerra, et al., Enzyme inhibition studies by integrated MichaelisMenten equation considering simultaneous presence of two inhibitors when one of them is a reaction product, Comput. Methods Programs Biomed. (2015),
http://dx.doi.org/10.1016/j.cmpb.2015.12.013
ARTICLE IN PRESS
c o m p u t e r m e t h o d s a n d p r o g r a m s i n b i o m e d i c i n e x x x ( 2 0 1 5 ) xxxxxx
The integrated MichaelisMenten equation in its classical implicit form [4,5] or in its explicit form [6] have other
advantages: (i) the initial addition of a product inhibitor can
be eliminated and Ki evaluated [710] allowing for example to
determine inhibition constants, of a particular reaction product isomer, more accurately since they are based on correct
stereochemistry and complete purity [11]; (ii) the possibility to
determine the inhibition constants without knowledge of true
substrate concentration [9,12,13]. This is of special mention in
heterogeneous reactions where the substrate is insoluble.
One of the main caveats regarding the use of integrated
MichaelisMenten equation to entire progress curve analysis is the possibility of gradual enzyme activity loss under
the assay conditions. However, nothing prevents its use with
much lower percent substrate depletion [2,5,10]. Thus, in this
work integrated equations will be applied only to data points
usually utilized to determine initial velocities (up to 10% substrate depletion).
The integrated MichaelisMenten equations have not been
usually applied in the presence of two inhibitors in the reaction medium. Nevertheless, integrated equations nd special
interest in studies with two different inhibitors since product inhibition is often found as a common regulating enzyme
mechanism [4,9,1417]. Thus, to study the inhibition kinetics
of this kind of enzymes it is necessary to consider the simultaneous presence of two inhibitors because one of them is a
reaction product (it is always present). The main objective of
this work is to develop and apply integrated equations to the
kinetic analysis of reactions carried out in the presence of two
inhibitors when one of them is a reaction product whether or
not mutually exclusive.
2.
2.1.
2.2.
3.
Theory
Kinetic analysis with the integrated MichaelisMenten equations permits the determination of inhibition constants (Ki )
when more than one reaction product inhibits the enzyme
[9,14,16]. In this situation:
n
1
j=1
Kij
1
Ki
where Kij is the inhibition constant for one single product and
Ki is the global inhibition constant taking into account all products. In enzyme reactions characterized by the presence of two
different product inhibitors, Hsu and Tsao [20] showed that if
there is no addition of an initial inhibitor (at time zero) the
two constants, Kij1 and Kij2 cannot be individually determined
because they are in a lumped form. Nevertheless in the present
study a different condition occurs since urea is not a reaction
product which means that inhibition constants for urea and
phosphate are not lumped [17]. Furthermore, Orsi and Tipton
[4] advocated that when two inhibitors exhibit different inhibition types, linearization of integrated equations (e.g. [14]) is
not appropriate because different kinetic effect of inhibitors is
indistinctive. The use of non-linear regression is a simple way
to overcome linearization weakness [9].
All common types of linear inhibitions are included in
linear mixed inhibition model (MI) since some inhibition
constants can tend to innity, meaning that they are irrelevant
giving rise to other linear models (see Supplement 1) [5,10].
Assuming the mixed linear inhibition (MI) the following
MichaelisMenten rate equation is obtained:
v=
Km 1 +
Vmax [S0 ]
[I]
Kic
+ [S] 1 +
[I]
Kiu
(1)
where Km , Michaelis constant; Kic , inhibitor dissociation constant of enzyme-inhibitor complex; Kiu , inhibitor dissociation
constant of enzyme-substrate-inhibitor complex; Vmax , maximum velocity of reaction; v, initial velocity of reaction; [I]
concentration of inhibitor and [S0 ] initial concentration of substrate.
Considering the presence of two different inhibitors I1 and
I2 as well as different inhibition constants Kic1 and Kic2 or Kiu1
and Kiu2 the rate equation becomes:
v=
Km 1 +
Vmax [S0 ]
[I1 ]
Kic1
[I2 ]
Kic2
+ [S] 1 +
[I1 ]
Kiu1
[I2 ]
Kiu2
(2)
Please cite this article in press as: R.M.F. Bezerra, et al., Enzyme inhibition studies by integrated MichaelisMenten equation considering simultaneous presence of two inhibitors when one of them is a reaction product, Comput. Methods Programs Biomed. (2015),
http://dx.doi.org/10.1016/j.cmpb.2015.12.013
ARTICLE IN PRESS
c o m p u t e r m e t h o d s a n d p r o g r a m s i n b i o m e d i c i n e x x x ( 2 0 1 5 ) xxxxxx
d[S]
dt
Vmax [S0 ]
Km 1 +
[I1,0 ]
Kic1
[Pt ]
Kic1
[I2 ]
Kic2
+ [S] 1 +
[I1,0 ]
[Pt ]
Kiu1 + Kiu1
[I2 ]
Kiu2
(3)
In Eq. (3), inhibition constants Kic1 and Kic2 or Kiu1 and Kiu2
are in a lumped form but after integration assuming that Pt
increases with time, the inhibition constants cease to be in a
lumped form [9,21].
Rearranging and integrating, gives:
Km 1 +
[I1,0 ]
Kic1
+[S0 ] 1 +
[St ]
dt =
[S0 ][St ]
Kic1
[I1,0 ]
Kiu1
[I2 ]
Kic2
[S0 ][St ]
Kiu1
[I2 ]
Kiu2
d[St ]
Vmax [S0 ]
[S0 ]
(4)
Or
dt Vmax = Km 1 +
0
[S0 ]
[I1,0 ]
[I2 ]
+
+
Kic1
Kic1
Kic2
Km
[S0 ]
[I1,0 ]
[I2 ]
1
+
+
+
Kiu1
Kiu1
Kiu1
Kiu2
1
Kiu1
[St ]
[S0 ]
1
d[St ]
[St ]
[St ]
d[St ]
[S0 ]
[St ]
[S]d[St ]
(5)
[S0 ]
t=
1
[E]kv
+ 1
+
1
2
Km 1 +
[S0 ]
[I1,0 ]
[I2 ]
+
+
Kic1
Kic1
Kic2
Km
[S0 ]
[I1,0 ]
[I2 ]
+
+
+
Kiu1
Kiu1
Kiu1
Kiu2
1
Kiu1
ln
[St ]
[S0 ]
([St ] [S0 ])
([St ]2 [S0 ]2 )
(6)
SSE
p
n
WI
CI
NCI
UCI
3976.3
2
112
1455.2
4
112
468.7
4
112
1019.0
4
112
MI
385.1
6
112
4.
Please cite this article in press as: R.M.F. Bezerra, et al., Enzyme inhibition studies by integrated MichaelisMenten equation considering simultaneous presence of two inhibitors when one of them is a reaction product, Comput. Methods Programs Biomed. (2015),
http://dx.doi.org/10.1016/j.cmpb.2015.12.013
ARTICLE IN PRESS
c o m p u t e r m e t h o d s a n d p r o g r a m s i n b i o m e d i c i n e x x x ( 2 0 1 5 ) xxxxxx
Table 2 Summary of model discrimination using Akaike information criterion methodology in the presence of
phosphate and urea.
Models A/B
SSEA
SSEB
WI/MI
NCI/MI
3976.3
468.7
385.1
385.1
112
112
PA +1
PB +1
AICA
AICB
AICcA
AICcB
Probability B correct
3
5
7
7
405.8
170.3
152.3
152.3
406.0
170.9
153.4
153.4
252.6
17.5
1.0
0.99984
Table 3 Summary of obtained constants with the best model (MI model).
MichaelisMenten
Km (M)
Kic1 (M)
Integrated equation
Initial velocities
41.4 1.2
58.3 2.1
7.0 1.5
Kiu1 (M)
Kic2 (M)
Kiu2 (M)
2.4 0.3
3.4 0.2
3.3 0.3
3.4 0.2
3.0 0.0
3.0 0.0
60
40
Residuals
Time (s)
50
30
20
15
30
45
60
-3
10
0
0
10
-6
Phosphate (M)
0.12
0.06
Residuals
0
0
300
600
900
1200
1500
Time (s)
0.5
1.5
2.5
-0.06
-0.12
Fig. 2 (A) Residual plot from data of Fig. 1A. (B) Residual
plot from data of Fig. 1B. The plot patterns emphasize the
absence of systematic errors.
SSE
p
n
1327.5
2
62
CI
83.1
3
62
NCI
UCI
120.2
3
62
114.8
3
62
MI
83.1
4
62
Please cite this article in press as: R.M.F. Bezerra, et al., Enzyme inhibition studies by integrated MichaelisMenten equation considering simultaneous presence of two inhibitors when one of them is a reaction product, Comput. Methods Programs Biomed. (2015),
http://dx.doi.org/10.1016/j.cmpb.2015.12.013
ARTICLE IN PRESS
c o m p u t e r m e t h o d s a n d p r o g r a m s i n b i o m e d i c i n e x x x ( 2 0 1 5 ) xxxxxx
Table 5 Summary of model discrimination using Akaike information criterion methodology assuming only the assays
in the absence of urea.
Models A/B
SSEA
SSEB
WI/CI
CI/MI
1327.5
83.1
83.1
83.1
n
62
62
PA +1
PB +1
AICA
AICB
AICcA
AICcB
3
4
4
5
196.0
26.1
26.1
28.1
1964.4
26.8
26.8
29.2
Changes to make
in the MI model
Number of
parameters
remaining
WIm
CIm
NCIm
UCIm
MI
3
3
3
4
SSE
p
n
WIm
CIm
NCIm
UCIm
4258.3
2
112
1456.0
3
112
402.3
3
112
1096.9
3
112
MIm
392.6
4
112
169.6
2.4
Probability B correct
1.00
0.23
product it is not possible to study the inhibition of urea without the presence of phosphate. Thus, it is necessary to copy
MI equation model (Supplement 1) to ve spreadsheets and
to give e.g. the following names to the ve spreadsheets
(WIm , CIm , NCIm , UCIm and MIm ; when m means modied).
Modications of MI model are necessary because phosphate
inhibition constants (Kic1 and Kiu1 ) are now xed values and
it is not possible to do that with the equations of Supplement 1. Thus Kic1 (cell B2) and Kiu1 (cell B5) will be xed by
inserting in each spreadsheet the values obtained in previous study without the presence of urea (Kic1 = 10.7 M and
Kiu1 = (e.g. 3.3 106 M obtained in Table 3), i.e. converting
them in non-variable parameters). Additionally, other minor
changes in each of the ve spreadsheets should be performed
as explained in Table 6. Before carried out nonlinear regression, do not forget to remove in the option by change the
cell (Solver dialog box) the cells $B$2 (Kic1 ) and $B$5 (Kiu1 ).
The discrimination result in Tables 7 and 8 show that urea
is a mixed inhibitor with a probability of 57%. The constants
obtained were Kic2 = 2.6 M and Kiu2 = 3.3 M.
Data points and model MI simulation with solver optimized constants (Table 3, integrated equation) are presented
in Fig. 1A. The constants obtained with the integrated
MichaelisMenten equation and by traditional methods (initial velocities) are compared in Table 3. It should be pointed
out that determination of inhibition constants, Kic1 and Kiu1 ,
by the MichaelisMenten equation (initial velocities) requires
the initial addition of phosphate in another set of assays
without urea. As above explained, the study of urea inhibition alone by initial velocities is not theoretically possible
because phosphate is a reaction product. Several authors concluded that urea is a noncompetitive inhibitor of alkaline
phosphatase [26,27]. Nevertheless, in this work, the probability of alkaline phosphatase inhibition by urea to be of the
noncompetitive type is not fully supported because slightly
different inhibition constants (Kic2 and Kiu2 ) were obtained
in the MI model discriminated (Table 8) by Akaike information criterion. This difference could be due to the fact that
effect of phosphate inhibition was not considered in traditional methods by initial velocities. Now it is possible to
state that phosphate is a competitive inhibitor and urea is
a mixed inhibitor with the inhibition constants presented in
Table 3.
It is important to point out that some alkaline phosphatase
isoenzymes present a reversible inhibition by urea (up to 2 M)
Table 8 Summary of model discrimination using Akaike information criterion methodology assuming the
modications of Table 6 (in the presence of phosphate and urea).
Models A/B
SSEA
SSEB
NCIm /MIm
402.3
392.6
112
PA +1
PB +1
AICA
AICB
AICcA
AICcB
150.5
148.4
151.6
151.1
0.54
Probability B correct
0.57
The subscript m in NCIm /MIm denotes that are modied MI equations as explained in text.
Please cite this article in press as: R.M.F. Bezerra, et al., Enzyme inhibition studies by integrated MichaelisMenten equation considering simultaneous presence of two inhibitors when one of them is a reaction product, Comput. Methods Programs Biomed. (2015),
http://dx.doi.org/10.1016/j.cmpb.2015.12.013
ARTICLE IN PRESS
c o m p u t e r m e t h o d s a n d p r o g r a m s i n b i o m e d i c i n e x x x ( 2 0 1 5 ) xxxxxx
and others are more sensitive and present an irreversible inhibition by denaturation [26,28], a phenomenon not observed
in this study. The use of the integrated MichaelisMenten
equation to the entire progress curve is known for long time.
The option to use only initial data range was chosen in order
to theoretically eliminate possible interferences such as,
enzyme denaturation or change of initial conditions (pH,
temperature, etc.).
5.
Conclusions
Conict of interest
None.
Acknowledgment
Authors are grateful for the nancial support by CITAB through
project FCOMP-01-0124-FEDER-022692.
references
Please cite this article in press as: R.M.F. Bezerra, et al., Enzyme inhibition studies by integrated MichaelisMenten equation considering simultaneous presence of two inhibitors when one of them is a reaction product, Comput. Methods Programs Biomed. (2015),
http://dx.doi.org/10.1016/j.cmpb.2015.12.013