Journal of Food Engineering 176 (2016) 121e127

Contents lists available at ScienceDirect

Journal of Food Engineering

journal homepage: www.elsevier.com/locate/jfoodeng

Analysis of the operational strategies for the enzymatic hydrolysis of

food proteins in batch reactor
P.L. Valencia a, *, S.A. Flores b, M.J. Pinto a, S.F. Almonacid a, c
a
Department of Chemical and Environmental Engineering, Universidad T
ecnica Federico Santa Mara, P.O. Box 110-V, Valparaso, Chile
b
Department of Mathematics, Universidad T
ecnica Federico Santa Mara, P.O. Box 110-V, Valparaso, Chile
c
Centro Regional de Estudios en Alimentos y Salud (CREAS), CONICYT-Regional, R12C1001 Valparaso, Chile

a r t i c l e i n f o a b s t r a c t

Article history: The aim of this work was to validate a mathematical model and evaluate hydrolysis of salmon muscle
Received 23 February 2015 proteins by Alcalase in batch reactor. The inhibition constants were correlated with conversion and
Received in revised form temperature thus improving dramatically the agreement between model predictions and experimental
2 September 2015
hydrolysis time courses. The standard operation in batch reactor consisting in pre-heating, isothermal
Accepted 22 October 2015
Available online 30 October 2015
stage and inactivation of enzyme was evaluated through simulation of different operational strategies
based on increasing temperature gradients. Operational conditions were 7.5%(w/w) protein, 1 g/l Alcalase
and different temperatures from 50
C to 73
C. The optimal temperature for the standard process was
Keywords:
Enzymatic hydrolysis
68
C with a processing time of 41 min. Different operational strategies, included temperature proles,
Protein hydrolysis resulted in longer operation times. The standard process was the best reactor performance due to the
Mathematical modeling maximization of the enzyme efciency.
Operational strategy 2015 Elsevier Ltd. All rights reserved.

1. Introduction reaction and inactivation stages. The model has to be sensitive to

Enzymatic hydrolysis of proteins is an attractive technology to
valorize food industry by-products and to produce protein hy-
drolysates with enhanced functional properties (Kristinsson and
Rasco, 2000a). The sh protein hydrolysates have been widely
characterized through their technological and bioactive proper-
ties (Kristinsson and Rasco, 2000a). Nevertheless, the enzymatic
hydrolysis of proteins has several disadvantages. The most
important are the high cost of enzymes and the need to inacti-
vate the enzyme by pH or heat treatment after the required
degree of hydrolysis is reached (Kristinsson and Rasco, 2000a).
The standard process for producing sh protein hydrolysates is
detailed in Fig. 1. In general, the process involves a reaction stage
followed by an inactivation stage in a batch reactor. Both the
reaction and inactivation stages require heat transfer. The reac-
tion stage is generally operated on isothermal condition at an
ideally optimal temperature; meanwhile the inactivation stage is
operated on temperatures ranging from 75
C to 100
C for
5e30 min (Kristinsson and Rasco, 2000b). A kinetic model is
needed to evaluate the batch reactor performance during both
* Corresponding author.
E-mail address: pedro.valencia@usm.cl (P.L. Valencia).
temperature in order to evaluate the effects on enzyme efciency
and inactivation rate. The kinetic of the enzymatic hydrolysis of
proteins has been modeled using the MichaeliseMenten equa-
tion for different enzymes and protein sources (Apar and Ozbek,
2009; Barros and Xavier Malcata, 2004; O'Meara and Munro,
1985; Trusek-Holownia, 2008). The inhibition has been
included in more complex models where the hydrolyzed proteins
and peptides inhibit the enzyme (Demirhan et al., 2011; Sousa
et al., 2004; Tardioli et al., 2005; Valencia et al., 2014). The re-
action system in the hydrolysis of proteins is very complex in
terms of substrate and products composition. The lack of a
proper denition of these variables has been a constant problem
in the cited articles. Nevertheless, during the study of the hy-
drolysis of salmon muscle proteins by Alcalase, Valencia et al.
(2014) solved this problem by dening the substrate concentra-
tion as the peptide bond concentration and the product con-
centration as the concentration of a-amino groups released
during hydrolysis. Both, the thermal inactivation of enzyme
during the hydrolysis and the effects of the modulation exerted
by the substrate and products on Alcalase, were also modeled by
Valencia et al. (2014). It was found that substrate and products
increased the thermal stability of Alcalase (positive modulation)
approximately 10 times even at substrate concentrations as low

http://dx.doi.org/10.1016/j.jfoodeng.2015.10.029
0260-8774/ 2015 Elsevier Ltd. All rights reserved.
122 P.L. Valencia et al. / Journal of Food Engineering 176 (2016) 121e127

Nomenclature KS Substrate uncompetitive inhibition constant (mM)

K1 Product competitive inhibition (mM)
S Substrate concentration (mM) K2 Product uncompetitive inhibition (mM)
E Enzyme concentration (g/l) Ki,0 Product inhibition constant independent of conversion
T Temperature (K) (mM)
S0 Initial substrate concentration (mM) b1 Slope of linear correlation between conversion and K1
E0 Initial enzyme concentration (g/l) (mM)
P Product concentration (mM) b2 Slope of linear correlation between conversion and K2
DH Degree of hydrolysis (mM)
X Degree of conversion M, N Slope and intercept of linear correlation between bn
t Reaction time (min) and T
kcat Catalytic constant (mM$l$g1$min1) T0 Initial temperature of operation (K)
kd Thermal inactivation constant (min1) m Temperature rate of T-proles (
C$min1)
K MichaeliseMenten constant for substrate (mM) a Exponential value of T-proles

as 15 mM of free amino groups from proteins. The kinetic model

developed by Valencia et al. (2014) considers the effect of sub- dS k T$E $ekd T$t $S
strate inhibition, product inhibition, modulated enzyme inacti-   cat  0   (1)
dt
vation and the temperature effect on the catalytic constant (kcat), KT 1 KP1 S 1 KP2 KSST
inactivation constant (kd), MichaeliseMenten constant (K) and
the substrate uncompetitive inhibition constant (KS). However, The kinetic, afnity and inhibition constants kcat, kd, K and KS are
the product inhibition constants, K1 and K2, were not correlated function of temperature according to Arrhenius and van't Hoff
with temperature. Additionally, the evidence indicated that K1
and K2 values change during hydrolysis reaction, which agrees
with the changing composition of hydrolysate. The challenge of
the present work is to determine the evolution of inhibition
constants during hydrolysis reaction and the dependence with
temperature to complete the model. Finally, the model will be
sensitive to changes in substrate and product concentration, and
changes in temperature. Those characteristics are needed to
evaluate the different operating strategies that involve temper-
ature gradients in a batch reactor. The problem of the present
work is to determine if the standard operation in batch reactor
results in the optimal reactor performance. The aim of this work
is to evaluate different operational strategies such as non-
isothermal operation with temperature prole. The comparison
of the different operational strategies will allow elucidating the
optimal reactor performance.

2. Batch reactor model

The kinetic model for the enzymatic hydrolysis of salmon

muscle proteins by Alcalase involves the variables substrate,
product and enzyme concentration. The substrate concentration (S)
was dened in terms of peptide bonds cleavable by Alcalase, which
corresponds to the 26.2% of the total available peptide bonds
(Valencia et al., 2014). The degree of hydrolysis (DH) of proteins is
the number of peptide bonds cleaved divided by the total number
of peptide bonds available in the reaction mixture. The degree of
conversion (X), is dened as the ratio between hydrolyzed peptide
bonds and those cleavable by Alcalase. A DH of 26.2% is equivalent
to the total conversion of available peptide bonds, when the degree
of conversion (X) is 1 (Valencia et al., 2014). The product concen-
tration (P) was dened in terms of the free a-amino groups
generated when a peptide bond is hydrolyzed. The enzyme con-
centration (E) corresponds to the mass of Alcalase per reaction
volume. The model considers uncompetitive substrate inhibition
and mixed product inhibition. The enzyme thermal inactivation
kinetics was modeled by a rst-order equation. The mass balance
for the batch reactor considering the kinetic model for the rate of
substrate conversion is presented in Eq. (1). Fig. 1. Standard process to produce sh protein hydrolysate.
 P.L. Valencia et al. / Journal of Food Engineering 176 (2016) 121e127 123

equations (Eqs. (2e5)). temperature until the desired value of residual activity is reached.
  Integration of Eq. (1) yielded Eq. (9), which was used to evaluate the
22:566801
conversion prole X vs t at a xed T.
T
kcat T e mM$l=g$min (2)    
S20 X 2 S X S S2 X2
  S0 X  K ln1  X 0 0 ln1  X 0 X 
2K2 K1 K1 KS 2
 
0:970$S0 11539746
T kcat $E0  
kd T 1 $e 1=min (3) 1  ekd t
2:21 S0 kd
  (9)
14:44045
T
Where X corresponds to the degree of conversion dened as P/S0.
KT e mM (4)
Eqs. (2e8) were used to include the T-prole during the inactiva-
  tion stage as follow:
6564
14:46
T T0 mt a
T
KS T e mM (5) (10)

Additionally, the thermal inactivation constant (kd) considers Where m is the T increase rate, T0 is the isothermal stage temper-
the effect of positive modulation by substrate and products through ature, a is an exponential factor and t is processing time. A
an empirical model developed from experimental data obtained by m 4
C=min and a 1 were chosen for the T-proles during pre-
Valencia et al. (2014). The inhibition constants (K1 and K2) deter- heating and inactivation stages. The residual activity (a) was
mined by Valencia et al. (2014) were 2.31 and 7.12 mM for a hy- dened according to rst-order inactivation model.
drolysate with a 5% degree of hydrolysis and 13.8 and 85.2 mM for a
hydrolysate with a 20% degree of hydrolysis. E
a ekd t (11)
E0
3. Methodology
The enzyme was added at time zero when the pre-heating stage
began. Two different initial temperatures were evaluated, an
3.1. Model validation
ambient temperature of 20
C and a cold temperature of 5
C. The
cold temperature was considered for simulating a refrigerated
Hydrolysis time courses at different initial substrate concen-
storing temperature of the substrate.
trations (S0), initial enzyme concentrations (E0) and temperatures
(T) were carried out. In a rst approach, the model predictions were
3.3. Simultaneous conversion-inactivation operation
tested on those experiments using xed values of inhibition con-
stants (K1, K2) obtained in Valencia et al. (2014). Hydrolysis time
The objective of this strategy is the simultaneous achievement
courses were carried out at different operational conditions of S0, E0
of both targeted operational parameters, X and a during the oper-
and T according to the methodology proposed by Valencia et al.
ation of the batch reactor. The major impact of this strategy is the
(2015). The degree of conversion against reaction time was used
intensication of the process due to the shift from multi-stage
to calculate inhibition constants values as functions of the degree of
(standard process) to one-stage operation. There are two possibil-
conversion (X), according to Eq. (6). The slope of the linear equation
ities to operate under this strategy: the isothermal and the T-prole
(bn) was correlated with temperature (Eq. (7)). Different Ki,0 values
operation.
were obtained at different temperatures. Analogue to K and KS,
inhibition constants were correlated with temperature using the
van't Hoff equation (Eq. (8)). 3.3.1. Isothermal operation
The isothermal operation consisted in the simultaneous reaction
Ki Ki;0 bn X (6) and inactivation processes at a temperature which allowed to reach
the desired X and a at the same time (t). This strategy requires a
pre-heating stage as in the standard process operation. The prob-
bn M$T N (7)
lem is formulated as follows. An objective function (F) was deduced
 
Table 1
DH
RT
lnK0
Operation conditions of the batch reactor for the enzymatic hydrolysis of proteins.
Ki;0 T e mM (8)
Variable Symbol Value Units
Where Ki,0 is the inhibition constant at temperature T, and M, N are Initial substrate concentration S0 165 mM
empirical constants for the correlation between bn and tempera- Initial enzyme concentration E0 1.00 g/l
ture. The complete model was used to predict the hydrolysis time Degree of conversion X 0.382 e
courses at different values of S0, E0 and T. Predictions were Residual activity a 0.010 e

compared with experimental data to validate the model.

Table 2
3.2. Standard process
Inhibition constants models as a function of temperature and conversion.

The standard process consisted on pre-heating, isothermal and Variable Value Units
 
inactivation stages. The pre-heating stage consisted in the tem- K1,0 mM
5343:4
T 16:25
perature rising of the batch reactor containing the substrate until e
the conversion temperature is reached. The isothermal stage is b1 0.0401$T11.82 mM
operated until the desired degree of conversion (X) is reached. The K2,0 190.0 mM
b2 185.8 mM
inactivation stage consisted in the progressive increasing of
124 P.L. Valencia et al. / Journal of Food Engineering 176 (2016) 121e127

Fig. 2. Experimental hydrolysis time courses (points) and model predictions with xed inhibition constants values (dashed lines) and varying inhibition constants (continuous
lines). Operating conditions (S0, E0 and T) A: 110 mM, 1.7 g/l, 60
C (circles); 77 mM, 0.83 g/l, 44
C (diamonds); B: 143 mM, 1.7 g/l, 50
C (triangles); 143 mM, 2.60 g/l, 56
C (squares).

Fig. 3. Experimental hydrolysis time courses (points) and model predictions with varying values of inhibition constants (continuous lines). Operating conditions (S0, E0, T) A:
143 mM, 6.25 g/l, 56
C (squares); 110 mM, 1.7 g/l, 60
C (circles); 110 mM, 1.7 g/l, 50
C (up triangles); 77 mM, 0.83 g/l, 56
C (down triangles); 77 mM, 0.83 g/l, 44
C; B: 165 mM,
1.7 g/l, 50
C (squares); 143 mM, 2.6 g/l, 44
C (circles); 110 mM, 1.7 g/l, 50
C (up triangles); 110 mM, 1.7 g/l, 40
C (down triangles); 110 mM, 0.23 g/l, 50
C.

Fig. 4. Conversion (X) and residual activity (a) proles during the standard process at 50
C (A), 60
C (B), 68
C (C) and 72
C (D) starting at 20
C. Dotted vertical lines mark the
limits of each stage.
 P.L. Valencia et al. / Journal of Food Engineering 176 (2016) 121e127 125

Table 3 Table 4
Reaction time (t) and conversion (X) during the standard process operation starting Reaction time (t) and conversion (X) during the standard process operation starting
at 20
C. at 5
C.

Isothermal stage T Pre-heating Isothermal Inactivation Isothermal stage T Pre-heating Isothermal Inactivation

t (min) X t (min) X t (min) X t (min) X t (min) X t (min) X

50 C 7.5 0.054 83.53 0.382 89.31 0.400 50 C 11.25 0.059 86.97 0.382 92.73 0.400



60 C 10.0 0.093 44.17 0.382 48.24 0.397 60 C 13.75 0.096 47.81 0.382 51.52 0.397



68 C 12.0 0.132 34.50 0.382 36.47 0.388 67 C 15.50 0.130 38.76 0.382 40.89 0.389



72 C 13.0 0.151 52.93 0.382 53.61 0.382 72 C 17.00 0.152 71.89 0.382 72.34 0.382

from Eq. (9) to calculate the T value to achieve the X and a simul- F(T) 0. The solution space contains a unique temperature value to
taneously at dened operational conditions (S0, E0). satisfy the specied constrains at xed values for the operating
conditions S0 and E0.
C kcat T$E0
FT A  B$KT  D (12)
KS T kd T 3.3.2. T-prole operation
As well as the one-stage isothermal operation, the T-prole
Where A, B, C, y D correspond to the following terms in Eq. (12): operation is proposed to achieve the X and a simultaneously in the
batch reactor. However, in this case the temperature is not constant
S20 X 2 but progressively increasing. The formulation of the problem is
A S0 X (13) based on the T-prole, where T is a function of operation time (t) as
2K2
dened in Eq. (10). Linear, convex and concave T-proles with a
values of 1, 2 and 0.5, respectively, were considered for evaluation.
S0 X S0
B ln1  X ln1  X (14) All the operation strategies were operated to achieve X 0.382,
K1 K1
corresponding to a DH 10%. The inactivation stages were carried
out until the 1% of the residual activity was left (a 0.01). The
 
X2 chosen initial substrate concentration was 165 mM of cleavable
C S20 X  (15)
2 peptide bonds equivalent to 7.5% (w/w) of proteins. Considering
that the salmon muscle contains 21% of proteins, the concentration
  of raw material inside the batch reactor was approximately 36%(w/
D 1  ekd t (16) w) of the reaction mixture. The initial enzyme concentration of 1 g/l
corresponds to an Alcalase activity of 2.4 AU/l of reaction mixture.
Where, the term D includes the residual activity (a) dened in Eq. The operation conditions were listed in Table 1. During the inacti-
(11). The problem consisted in computing the T value to make vation stage the temperature was increased until 85
C and then

Fig. 5. Conversion (X) and residual activity (a) proles during the standard process at 50
C (A), 60
C (B), 67
C (C) and 72
C (D) starting at 5
C. Dotted vertical lines mark the
limits of each stage.
126 P.L. Valencia et al. / Journal of Food Engineering 176 (2016) 121e127
Fig. 6. Conversion (X) and residual activity (a) proles during the simultaneous
conversion-inactivation operation starting at 5
C and the isothermal stage operated at
72.3
C. Dotted vertical lines mark the limits of each stage.
Table 5
T-function parameters for the different T-prole strategies starting at 5
C.
Strategy m a
One-stage linear T-prole 0.655
One-stage convex T-prole 0.00191
One-stage concave T-prole 8.656
maintained until a 0.01.
4. Results and discussion
A kinetic model for the enzymatic hydrolysis of proteins ob-
tained in a previous article modeled the effect of substrate con-
centration, enzyme concentration and temperature (Valencia et al.
2014). Kinetic constants kcat, kd, K, KS were correlated with tem-
perature. The mixed product inhibition was characterized through
the inhibition constants K1 and K2. Results indicated that these
constants change during the progress of hydrolysis reaction. The
model was used to predict the experimental hydrolysis time
courses using xed values of K1 and K2. In the present article, both
inhibition constants were correlated with temperature and degree
of conversion (X) to model their evolution during the hydrolysis
reaction. The results obtained were listed in Table 2. The inhibition
constant K1 varied as a function of temperature and conversion
(Eqs. (6e8)) meanwhile K2 did not correlated with temperature and
a constant value of 190.0 mM was obtained. However, K2 did
correlate with conversion according to Eq. (6). The results also
indicated that empirical constant b1 was correlated with temper-
ature meanwhile b2 was not, as well as occurred with K1 and K2
Fig. 7. Conversion (X) and residual activity (a) proles during the one-stage T-proles operation starting at 5
C for linear (A), concave (B) and convex (C) T-proles. Dotted vertical
line marks the end of operation.
respectively. A selection of experiments and model predictions
both, using xed values of inhibition constants (K1 2.31 mM and
K2 7.12 mM) and varying values according to Eqs. (6e8) and
Table 2, were plotted in Fig. 2. The xed Ki values prediction agreed
with experiments in the initial phase of the hydrolysis time courses
until 30 min of reaction approximately. After that, an over-
estimation of the product concentrations during the hydrolysis
reactions was obtained. The lack of agreement between model
prediction and experimental data was supposed to be a conse-
quence of the variation of the inhibition constants values during
the hydrolysis progress. Predictions of hydrolysis time courses
including the variation of inhibition constants during reaction
progress (Table 2) was made to test that hypothesis. A very good
correction was made using the varying values of inhibition con-
stants as a function of temperature and conversion (continuous
lines in Fig. 2). Additional experimental data and model predictions
were plotted in Fig. 3. The model predictions agreed with experi-
mental data for different values of S0, E0 and T ranging from 3.5%(w/
w) to 7.5%(w/w), 0.23 g/l to 2.6 g/l and 40
Ce60
C, respectively.
The results endorsed the present model especially for the variation
of the inhibition constants with conversion and temperature. The
effects of temperature on the kinetic constants kcat and kd, and the
thermodynamic constants K, KS and K1 (K2 and b2 resulted unaf-
fected by T) were all nally included in a conceptual model to
1 evaluate different operational conditions and strategies. Optimi-
2 zation of the process can now be theoretically calculated.
0.5 The standard process for the production of protein hydrolysates
is carried out in three stages: pre-heating, isothermal and inacti-
vation. The operation proles of the batch reactor starting from
20
C with an isothermal stage at 50
C, 60
C, 68
C and 72
C were
plotted in Fig. 4. These results showed that the higher the operating
T the shorter the inactivation stage. The lower residual activity
remaining at the beginning of the inactivation stage caused this
effect. During the inactivation stage, the remaining active mole-
cules of enzyme continued hydrolyzing proteins, thus the nal
degree of conversion was slightly higher than 0.382, except at 72
C.
The difference is lower at higher T, as can be observed in detail in
Table 3. The optimal T for this operating strategy at conditions
indicated in Table 1 is 68
C. Previous reports have shown optimal
temperatures between 60
C and 70
C for the enzymatic hydrolysis
of proteins (Prieto et al., 2008).
Another strategy was evaluated within the standard process
consisting in the operation at a cold starting temperature of 5
C
simulating the storage of the substrate maintained under refriger-
ated conditions before use. The process was evaluated at a starting
temperature of 5
C with the pre-heating followed by the
isothermal and inactivation stages, as plotted in Fig. 5. As a
consequence of the lower initial temperature, the processing times
were longer than those starting at 20
C (Table 4). The optimal
process was obtained at 67
C, corresponding to the shortest pro-
cessing time.
 P.L. Valencia et al. / Journal of Food Engineering 176 (2016) 121e127
Table 6
Processing times (t) for the different operation strategies starting at 5
C.
Strategy
Standard process
Simultaneous isothermal
One-stage linear T-prole
One-stage convex T-prole
One-stage concave T-prole
By analyzing the operations at 72
C in Figs. 4De5D, it can be
intuitively inferred that there is a trend to the simultaneous
achievement of X and a at some temperature over 72
C. Precisely,
the simultaneous conversion-inactivation operation consisted in
the simultaneous achievement of X and a. This strategy was eval-
uated at a temperature which allowed to accomplish the set values
of X and a. The results for the simultaneous conversion-inactivation
operation are plotted in Fig. 6. The constrains were accomplished at
T 72.3
C. Nevertheless, the performance of the batch reactor was
lower than the optimal performance in the standard process. The
operation time was 120 min. The temperature for the simultaneous
conversion-inactivation operation depends on the operation con-
ditions. A different temperature would be obtained at different
initial substrate concentration (S0), enzyme concentration (E0) or
degree of conversion (X).
Another option to achieve simultaneously the conversion and
inactivation degrees was evaluated. It consisted in the T-prole
operation where the temperature was progressively increased
until the objective values of X and a were reached. The most
relevant characteristic of this operation strategy was the one-
stage operation due to the unnecessary pre-heating and inacti-
vation stages. The linear T-prole was determined calculating
the value of m in Eq. (10) with a 1. The same method was
followed for the concave and convex T-proles with a 0.5 and
a 2 respectively. The results for the T-function values were
listed in Table 5. The results for all the T-proles starting at 5
C
are shown in Fig. 7. The reactor performance was not better than
the standard operation with processing times ranging from 103
to 186 min (Table 6). The concave T-prole was the most pro-
ductive from all the T-proles strategies (Table 6). However, the
concave T-prole resulted in 64 min operation compared to
41 min obtained in the standard operation (Table 6). On the
other hand, it can be observed that the performance obtained
with the concave T-prole approaches the one used in the
standard process. Finally, according to the operation times ob-
tained (Table 6), it can be concluded that the standard operation
was the most productive from all the evaluated strategies.
Despite the fact that this model was developed for the experi-
mental system salmon muscle proteins/Alcalase, it can be
extrapolated that the standard operation will be the optimal
operational strategies for other systems protein/protease. At
least, a new operational hypothesis, based on experimentally
validated kinetic models, has been formulated, which can be
tested by empirical data regarding operational strategies of the
enzymatic batch reactor.
127
5. Conclusions
t (min) A conceptual model considering the effect of temperature on all
41
the kinetic and thermodynamic constants was obtained. The
120 incorporation of inhibition constants depending on the degree of
103 conversion and temperature during the progress of hydrolysis re-
186 action increased dramatically the agreement between the model
64
predictions and experimental data. The simulation of the batch
reactor performance for the hydrolysis of salmon muscle proteins
by Alcalase with different strategies was achieved. It was demon-
strated that the standard operation consisting in pre-heating,
isothermal and inactivation stages was the most productive strat-
egy to enzymatically hydrolyze the salmon muscle proteins with
Alcalase in a batch reactor. The isothermal stage involved in the
standard operation was key to optimize the process because it
allowed obtaining the maximum efciency of the enzyme.
Acknowledgments
The authors are grateful for the nancial support provided by
CONICYT, FONDECYT/Initiation 11110249 and FONDECYT/Regular
1121147.
References
Apar, D.K., Ozbek, B., 2009. A kinetic study on corn gluten hydrolysis. Chem. Eng.
Technol. 32, 673e675. http://dx.doi.org/10.1002/ceat.200800468.
Barros, R.M., Xavier Malcata, F., 2004. A kinetic model for hydrolysis of whey pro-
teins by cardosin A extracted from Cynara cardunculus. Food Chem. 88,
351e359. http://dx.doi.org/10.1016/j.foodchem.2004.01.046.
Demirhan, E., Apar, D.K., Ozbek, B., 2011. A kinetic study on sesame cake protein
hydrolysis by Alcalase. J. Food Sci. 76, C64eC67. http://dx.doi.org/10.1111/j.1750-
3841.2010.01938.x.
Kristinsson, H.G., Rasco, B.A., 2000a. Fish protein hydrolysates: production,
biochemical, and functional properties. Crit. Rev. Food Sci. Nutr. 40, 43e81.
http://dx.doi.org/10.1080/10408690091189266.
Kristinsson, H.G., Rasco, B.A., 2000b. Kinetics of the hydrolysis of Atlantic salmon
(Salmo salar) muscle proteins by alkaline proteases and a visceral serine pro-
tease mixture. Process Biochem. 36, 131e139. http://dx.doi.org/10.1016/S0032-
9592(00)00195-3.
O'Meara, G.M., Munro, P.A., 1985. Kinetics of the hydrolysis of lean meat protein by
alcalase: derivation of two alternative rate equations and their t to experi-
mental data. Biotechnol. Bioeng. 27, 861e869. http://dx.doi.org/10.1002/
bit.260270616.
Prieto, C.A., Guadix, E.M., Guadix, A., 2008. Inuence of temperature on protein
hydrolysis in a cyclic batch enzyme membrane reactor. Biochem. Eng. J. 42,
217e223. http://dx.doi.org/10.1016/j.bej.2008.06.018.
Sousa, R., Lopes, G.P., Tardioli, P.W., Giordano, R.L.C., Almeida, P.I.F., Giordano, R.C.,
2004. Kinetic model for whey protein hydrolysis by alcalase multipoint-
immobilized on agarose gel particles. Braz. J. Chem. Eng. 21, 147e153. http://
dx.doi.org/10.1590/S0104-66322004000200003.
Tardioli, P.W., Sousa, R., Giordano, R.C., Giordano, R.L.C., 2005. Kinetic model of the
hydrolysis of polypeptides catalyzed by Alcalase immobilized on 10% glyoxyl-
agarose. Enzyme Microb. Technol. 36, 555e564. http://dx.doi.org/10.1016/
j.enzmictec.2004.12.002.
Trusek-Holownia, A., 2008. Production of protein hydrolysates in an enzymatic
membrane reactor. Biochem. Eng. J. 39, 221e229. http://dx.doi.org/10.1016/
j.bej.2007.09.010.
Valencia, P., Pinto, M., Almonacid, S., 2014. Identication of the key mechanisms
involved in the hydrolysis of sh protein by Alcalase. Process Biochem. 49,
258e264. http://dx.doi.org/10.1016/j.procbio.2013.11.012.
Valencia, P., Espinoza, K., Ceballos, A., Pinto, M., Almonacid, S., 2015. Novel modeling
methodology for the characterization of enzymatic hydrolysis of proteins.
Process Biochem. 50, 1e9. http://dx.doi.org/10.1016/j.procbio.2014.12.028.