Biotechnology Reports 30 (2021) e00611

Contents lists available at ScienceDirect

Biotechnology Reports
journal homepage: www.elsevier.com/locate/btre

Optimization of enzymatic hydrolysis of red tilapia scales

(Oreochromis sp.) to obtain bioactive peptides
Leidy Maritza Sierra-Lopera, Jose Edgar Zapata-Montoya*
University of Antioquia, Nutrition and Food Technology Group, 70th Street No. 52 - 21, 050010, Medellin, Antioquia, Colombia

A R T I C L E I N F O A B S T R A C T

Article history: The objective of this study was to optimize the conditions of enzymatic hydrolysis (type of enzyme, pH,
Received 22 August 2020 temperature (T), substrate (S) and enzyme concentration (E)) to increase content of soluble peptides (P),
Received in revised form 6 February 2021 antioxidant activities and degree of hydrolysis DH (%), in hydrolysates. Also, the effect of scaling up from a
Accepted 19 March 2021
0.5 L to a 7.5 L reactor, was evaluated. Hydrolysis was carried out for 3 h in a 500 mL reactor, with
Alcalase1 2.4 L and Flavourzyme1 500 L enzymes. A second experimental design was then developed
Keywords: with S and E as factors, where DH, P and antioxidant activity, were response variables. The Alcalase1 2.4 L
Enzymatic hydrolysis
was the most productive enzyme, with optimal S and E of 45 g/L and 4.4 g/L, respectively. Its hydrolysates
Alcalase
Flavourzyme
showed antioxidant activities with IC50 of 0.76 g/L, 12 g/L and 8 g/L for ABTS, FRAP and ICA, respectively.
Antioxidant activity The scale up didn’t showed negative effect on the hydrolysis.
Red tilapia © 2021 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND
license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction Flavourzyme1 500 L, properase E, pronase, collagenase, bromelain,

or papain, are usually employed in these processes [6]. Enzymatic
In recent years, the fish industry has shown significant growth hydrolysis allows for the possibility of obtaining peptides
globally. Colombia is the tenth producer of red tilapia in the world, exhibiting biological activity, such as antioxidants, antihyperten-
which is a species that represents 62 % of the country’s fish sives, antimicrobials, anticoagulants [7], and compounds with
production [1]. 60 % of red tilapia production ends up as waste, calcium binding properties [7] or antitumor properties [8].
such as fillet remains (15–20 %), skin and fins (1–3 %), bones (9–15 An important parameter for hydrolysis is the type of enzyme
%), heads (9–12 %), viscera (12–18 %) and scales (5%) [2]. These high used, because it influences the sequence of the peptides which are
volumes of waste represent a challenge from an environmental generated [9]. One of the most used enzymes is Alcalase1 , which
standpoint, because approximately 50 % of the waste is discarded produces peptides exhibiting antioxidant [10], anticoagulant [11],
without being used and only 30 % is used for products with a lower antihypertensive [12] and calcium binding properties [13].
added value [3]. However, this waste is also an opportunity from an Another interesting enzyme is Flavourzyme1 500 L, which
economic standpoint, due to the fact that some of this waste has produces peptides capable of antioxidant activity [14] and
significant levels of protein. This is the case of red tilapia scales; exhibiting calcium binding properties [15]. Despite the favorable
whose protein content ranges between 41 % and 84 % - the highest results available in the literature, there are few studies that
among residues of fish farms [4]. The protein is insoluble and found evaluate obtaining biologically active peptides from red tilapia
within a solid matrix, and enzymatic hydrolysis is used as a scales (Oreochromis sp.) at different scales of production. Scaling
procedure to ensure its solubilization in the liquid medium to processes is a tool that allows to carry out laboratory scale
avoid alkaline or acid hydrolysis which produces undesirable processes at a greater industrial level. This discipline allows
byproducts. For this reason, enzymatic hydrolysis is one of the researchers to evaluate aspects such as energy consumption,
alternatives with the greatest potential for protein utilization, productivity, efficiency, purification, and separation of products,
which can improve the nutritional quality of the substrate and among many others. It is usually based on criteria such as similar
favor the release of bioactive compounds [5]. Commercial geometry, dynamics and dimensional analysis [16].
proteases such as trypsin, chymotrypsin, pepsin, Alcalase1 , The objective of this study was to optimize the conditions of
enzymatic hydrolysis by varying the type of enzyme, pH,
temperature, substrate, and enzyme concentration, to obtain
* Corresponding author.
protein hydrolysates exhibiting of antioxidant activities, as well as
E-mail addresses: [email protected] (L.M. Sierra-Lopera), evaluating the effect of scaling the reaction up from a 0.5 L to a 7.5 L
[email protected] (J.E. Zapata-Montoya). reactor.

https://doi.org/10.1016/j.btre.2021.e00611
2215-017X/© 2021 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
L.M. Sierra-Lopera and J.E. Zapata-Montoya Biotechnology Reports 30 (2021) e00611

2. Material and methods Switzerland), which was operated via a computer (tiamo software
1.2.1). The reaction system was maintained at a constant stirring of
2.1. Chemical and reagents 560 rpm using a 801 magnetic stirrer for 3 h (Metrohm,
Switzerland) [1]. The hydrolysis reaction was monitored based
The reagents 2,20 -Azino-bis (3-ethylbenzothiazolin)-6-sulfonic on the degree of hydrolysis (DH), because an increase in DH has
acid (ABTS), 6-hydroxy-2,5,7,8-tetramethylchromo-2-carboxylic been associated with a greater presence of low-molecular-weight
acid (Trolox), 3- (2-Pyridyl), 5–6 diphenyl acid monosodium salt peptides, which in turn are attributed to exhibit greater biological
hydrate (Ferrozine) were obtained from Sigma-Aldrich (Oakville, activity [23]. The DH was expressed as the ratio between the
Ontario, Canada) and 2,4,6-Tri-2-pyridyl-s-triazine (TPTZ) was number of hydrolyzed peptide bonds (h) and the number of total
supplied by Merck (Darmstadt, Germany). Alcalase1 2.4 L (com- peptide bonds in the native protein per unit weight (hTOT) [21]. The
mercial protease obtained from the fermentation of Bacillus DH was calculated with Eq. 1, according to the pH-stat method [21].
licheniformis, non-specific serine endopeptidase) was supplied
by Novozymes (Bagsværd, Denmark) and Flavourzyme1 500 L B NB
DH ð%Þ ¼      100 ð1Þ
(complex of fungal peptidases obtained from fermentation of Mp a hTOT
Aspergillus oryzae, endo and exo peptidase) was provided by Sigma-
Where B is the consumed volume of NaOH in L, MP is the mass of
Aldrich, (Oakville, Ontario, Canada). All the reagents used in the
the protein in kg, NB is the concentration of NaOH, and α is the
study were analytical grade.
degree of dissociation of the amino groups released during the
reaction. α and pK were calculated using Eqs. 2 and 3, respectively
2.2. Fish scales
[22]. The total number of peptide bonds (hTOT) was obtained from
the aminogram using Eq. 4.
The fish scales were supplied by Piscícola El Gaitero in San
Jerónimo-Colombia. They were washed and disinfected with
10pHpK
0.2 mg/L sodium hypochlorite, to be able to store them in /¼ ð2Þ
polyethylene bags at -20  C until their use. They were then ð1 þ 10pHpK Þ
dried at 60  C for 24 h, and their size was reduced to 850 mm, by
means of an endless screw mill (Corona, Colombia).
ð298  TÞ
pK ¼ 7:8 þ 2400 ð3Þ
2.3. Characterization of the scales 298T
According to Adler-Nissen, 1979 [19], it is possible to calculate
The proximal composition (moisture, protein, fat and ash) of the hTOT from the concentration of each of the amino acids present in
scales was determined using the AOAC 930.15, 990.03, 920.38, the scales using Eqs. 4 and 5.
942.05 methods, respectively [17].
hTOT¼  1 ð1þFiÞ1000 ð4Þ
P
18
Fi
2.4. Amino acid analysis i¼1

The ground scales were subjected to an acid hydrolysis using

6 N HCl and 0.1 % phenol. Hydrolysis was carried out at 110  C for ½Ai
24 h. The derivatization of the samples was done before the pre- Fi ¼ P18 ðPMÞi ð5Þ
i¼1 ½Ai
column, using o-phthalaldehyde (OPA) for primary AA and 9-
fluorenyl methyl chloroformate (FMOC) for secondary AA [18]. The Where the amino acid is i, Ai is the concentration (mmol/mL), ðPMÞi
amino acids were analyzed using a HPLC Ultimate 3000 (Thermo is the molecular weight of amino acid i, and Fi is the molar fraction
Fisher Scientific, United States), with an UV/VIS detector with a of amino acid i. The hTOT was calculated using the amino acid
DAD diode array, operated with a 5 mm analytical column, ZORBAX serine.
Eclipse AAA-C18, 4.6  75 mm (Agilent, United States). A mobile At the end of the hydrolysis, the enzymes were inactivated at
phase in the form of a gradient prepared by a quaternary pump was 90  C for 10 min. The resulting hydrolysate was centrifuged at
used to mix the water, methanol, acetonitrile and 40 mM Buffer 1160 g for 20 min at 20  C and the supernatant containing small
NaH2PO4 solution with a pH of 7.8 [18]. size peptides was collected and stored at 80  C until use.

2.5. Enzymatic hydrolysis 2.6. Antioxidant activity

First, the protein content of the raw material was determined by 2.6.1. ABTS free radical cation scavenging assay
Kjeldahl [17] using a 6.25 conversion factor. The content of soluble The ABTS solution was prepared based on Zheng, Zhao, Xiao,
peptides from the hydrolysates were determined based on Biuret Zhao, & Su, 2016 [25], and was mixed with 7 mM ABTS+ and
[24] using the bovine serum albumine calibration curve from MP 2.45 mM potassium persulfate. Then, 1 mL of the solution was
Biomedicals (Shandong, China) as a reference. added to the 100 mL of sample or Trolox and left in the dark for 1 h.
No previous protein release processes were carried out, due to The absorbance was measured at 730 nm. The results were
the fact that these can generate negative effects on the structure of calculated using a standard Trolox curve and expressed as
the protein and therefore on its biological activities [20]. For this micromoles of Trolox equivalent (mmol TE/L).
reason, the enzymatic hydrolysis was established the first process
in the experiment. 2.6.2. Ferric-reducing antioxidant power (FRAP) assay
A glass reactor with a water circulation jacket was used for The FRAP was determined according to Pulido et al., 2000 [26].
temperature regulation, the reactor had a volumetric capacity of The FRAP solution was prepared with TPTZ 10 mM in HCl 40 mM,
1 L with a 500 mL work volume. The pH control and temperature FeCl3 20 mM and an acetate buffer 0.3 mM in the dark at 37  C.
recording were performed using a combined LL glass electrode 30 mL of sample or Trolox, 90 mL of distilled water, and 900 mL of
with a fixed frosted diaphragm (temperature between 080  C), the solution were added and the mix was left in the dark for 1.5 h.
connected to an automatic titrator (Titrando 842) (Metrohm, The absorbance was measured at 595 nm. The results were

2
L.M. Sierra-Lopera and J.E. Zapata-Montoya Biotechnology Reports 30 (2021) e00611

calculated using a standard Trolox curve and expressed as repetitions of the central point. The levels of S and E were
equivalent micromoles of Trolox (mmol TE/L). comprised within the range 8.845 g/L and 0.4–5.6 g/L respec-
tively. The response variables were DH (%), content in soluble
2.6.3. Iron chelating activity (ICA) peptides (g/L), ABTS (mmol TE/L), and ICA (%). The results were
Ferrozine forms complexes with ferrous iron producing a red analyzed using Design Expert software (version 7.0, Stat Easy
colour, and the chelating agents as peptides inhibit the formation Inc., Minneapolis, USA). Due to the relationship between DH and
of the complex, which leads to a decrease in the red colour. Thus, the antioxidant activity, this variable was included in the first
the measure of colour reduction is an estimation of the binding design as a response variable, but in final design, the antioxidant
ability of the chelator [27]. The test was performed following the activity was used directly, since it was the main objective of the
methodology of Choonpicharn et al., 2015, in which 1 mL of sample study.
or water (blank) is mixed with 20 mL of ferrozine and 40 mL of
ferrous sulfate, and left in the dark for 10 min. The absorbance was 2.7.4. Determination of the time of the enzymatic hydrolysis
measured at 562 nm, and the iron chelating activity (ICA) was Once the optimal hydrolysis conditions were found, P and the
calculated as shown in Eq. 6. ABTS, ORAC and ICA bioactivities were evaluated at different DHs
such as 2, 7, 12, 17 and 18, which relate to different hydrolysis times.
Ablank  Asample
ICA% ¼ x100 ð6Þ
Ablank
2.8. Enzymatic hydrolysis scale up

2.7. Optimization of the enzymatic reaction conditions for bioactive The objective of scaling up the reaction was to evaluate the
peptide production enzymatic hydrolysis in an industrial scale reactor with a similar
geometry, in order determine if the product could be further
2.7.1. The optimal pH and temperature conditions applied in the pharmaceutical and food industries.
The pH and temperature conditions depend of the substrate as After the optimization of the conditions in the 500 mL reactor,
reported by Hamid et al., 2015 [28]. For this reason, a central the process was transferred to a 7.5 L capacity reactor – Bioflo1 &
composite design was created for each enzyme (Flavourzyme1 Celligen 310, G628011 (New Brunswick Scientific, USA), stirred
500 L and Alcalase1 2.4 L) maintaining the same units of catalytic through a Rushton turbine with a working volume of 5 L (Edinson,
activity/g of substrate in both cases. This was done in order to NJ, USA) and fitted with temperature control. Optimal conditions
evaluate the effect of pH and temperature on two response used at this scale were 58.5  C, an enzyme-substrate ratio [E/S] of
variables: DH (%) and content of soluble peptides (P, g/L), thus 0.098 g enzyme/g protein and a pH of 8.05. The pH was kept
being able to determine the maximum soluble peptides in a liquid constant with the addition of 2 M of NaOH using an automatically
medium with a higher DH. controlled peristaltic waterfall pump and a pH sensor. The reactor
The substrate concentration refers to the initial content of protein was equipped with heating by using a recirculation system, it had
in the solid matrix in the reaction, and was kept constant at 8 g/L. The four baffles on the tank wall and a Rushton turbine impeller with a
enzyme was used at a 20 % w/w based on the substrate. The results diameter of 0.077 m. The dynamic similarity principle was used
were analyzed using version 7.0 of Design Expert software, (Stat Easy keeping the Reynolds number (Re) constant between the model
Inc., Minneapolis, USA). The pH levels for Alcalase1 2.4 L were (m) (500 mL) and the prototype (p) (7.5 L). Eq. 8 was used to
between 7.5–10.4 and betwen 7–8.5 for Flavourzyme1 500 L, and the calculate the velocity of the agitation in the prototype (p) where D
temperatures were between 40.9  C and 69.1  C for both enzymes.13 is the impeller diameter, r is the density, m is the viscosity and N is
random runs, 8 design points and 5 repetitions of the center point the velocity of agitation [31].
were carried out for each design. After the optimization of the ! !
models, the antioxidant and iron chelation bioactivities were N: D2i :r N: D2i :r
Rem ¼Rep !   ¼  ð8Þ
evaluated at the optimum point of each of the enzymes, which m m
m p
was then used to calculate the productivity.
1
In this model, N = 9.8 s and Di=5 cm, while the prototype
2.7.2. Determination of kinetic parameters Di=7.7 cm. Clearing N of the protype from Eq. (8) gives a velocity of
A saturation curve was derived for Alcalase1 2.4 L by varying the 250 rpm for the impeller in the prototype. However, to evaluate the
concentration of the substrate (S) from 4 g/L to 36 g/L, by means of effect of N on the hydrolysis process, a velocity of 350 rpm was also
hydrolysis carried out at the optimum pH and temperature evaluated in the prototype.
conditions determined during the previous stage. The enzyme
volume (695 mL) and agitation (560 rpm) were constant. The 2.9. Statistical analysis
experiments were evaluated for over 5 min and measurements
were taken every 10 s. Each curve enables the determination of the The measurements were taken in triplicate. All statistical
initial velocity (V 0 ), from which the multiplicative inverse (1/V 0 ) results were analyzed with a 95 % confidence level (p value <0.05).
was plotted against the multiplicative inverse of the substrate (1/S)
and the kinetic parameters maximum velocity (Vmax) and the 3. Results and discussion
Michaelis–Menten constant (Km) were obtained following the
Lineweaver-Burk model as shown in Eq. 7 [30]. 3.1. Characterization of the scales
 
1 Km 1 1
¼ þ ð7Þ Red tilapia scales (Oreochromis sp.) contained in dry basis
V 0 V max S V max
0.9  0.005 % fat, 83.9  1.04 % protein and 15.1  0.003 % ash.
These protein values show the by-product's potential for using the
2.7.3. Optimization of the substrate and enzyme concentration substrate as a source of protein. Other authors have shown lower
After acquiring the kinetic parameters of the reaction, it was values for protein, 49.42 %, in red tilapia scales (Oreochromis sp.)
possible to define the values of the substrate (S) and enzyme (E) [4] and 45.2 % in golden goatfish scales (Parupeneus cyclostomus)
concentration factors for a central composite design with 5 [32].

3
L.M. Sierra-Lopera and J.E. Zapata-Montoya Biotechnology Reports 30 (2021) e00611

3.2. Amino acid composition 3.3. Determination of hTOT

Table 1 shows the amino acid composition of red tilapia scales According to the aminogram shown in Table 1, the hTOT was
and hydrolysate. The table shows that the most abundant amino calculated using the amino acid lysine as a reference. The value
acid is glycine, followed by proline, glutamic acid and hydroxy- found was of 8.85, which is very close to the value reported by
proline, which were reported to be abundant in scales from nile Adler-Nissen 1986, who reported a value of 8.6, for fish in general.
tilapia (Oreochromis niloticus) [33], tilapia (Oreochromis sp.) [4]. This result is closer to the theoretical one, compared to the 9.3
However, alanine was found in a smaller proportion compared to value reported for by-products of Atlantic Salmon (Salmo salar)
the results of the mentioned authors. The amino acid composition [39].
refers to the types of proteins present in the scales which are type I
collagen, keratin and mucin [34]. High proline and glycine content 3.4. Optimization of enzymatic conditions for bioactive peptide
are due to the collagen present [35]. In the case of keratin, the production
predominant amino acids are glycine and glutamic acid [35].
Finally, mucin is a glycoprotein rich in serine, threonine and 3.4.1. Optimization of pH and temperature conditions
proline [36], which are also found in high proportions in the scales. An analysis of variance (ANOVA) was carried out to establish the
The presence of essential amino acids (9.5 %) is indicative of the effect of pH and T on the DH and P which is shown in Table 2. The
nutritional possibilities of protein in red tilapia scales (Table 1). On models have acceptable R2 values and a non-significant lack of fit
the other hand, the content and sequentiation of amino acids adjustment for all cases. The two factors have significant effects on
within the peptides is very important to explain their possible both responses for both enzymes. pH exerts a significant effect on
antioxidant activity. For example, the hydrophobic properties of DH, both in linear and quadratic terms when using Alcalase1 , in
amino acids can favor their interaction with the target lipids or the addition to T and pH interactions. However, in the case of
entry of the peptides into the target organs through an interaction Flavourzyme1 500 L, these factors only have a linear effect. On the
with the lipid bilayer in the cell, which is an important factor to other hand, P is affected linearly by both variables in Alcalase1 ,
achieve the desired antioxidant effects [37]. However, P is while in Flavourzyme1 500 L only temperature has an linear and
independent from the scales, and could be associated to the quadratic effect.
amino acid composition presented in Table 1. Accordingly, the Second order polynomial models were obtained from the
presence of aromatic amino acids (5%) and His, Pro, Met, Lys and ANOVA (Table 2) for each response in each enzyme. The
Cys, could be related to an increase in antioxidant activity because coefficients were then calculated by multiple regression. The
of their ability to donate protons or electrons [38]. models are the result of the exclusion of non-significant terms
In relation to the hydrolysate, the study found that the (p > 0.05), thus preserving the hierarchy of the model.
proportion of aromatic and non-essential amino acids increased
with respect to the raw material. Which entails that they are more 3.4.1.1. Effects of pH and temperature on Alcalase1 2.4 L. In the case
chemically available to be quantified after the hydrolysis processes. of Alcalase1 , Eqs. 9 and 10 describe the behavior of DH and P,
This is not the case for hydrophobic and essential amino acids, respectively. The differences in signs between the linear and
which decreased in the hydrolysate due to the hydrolysis quadratic terms in the same equation indicate that there is an
conditions of the extraction method. extreme point for the given response in the working range
depending on the variable in question. For DH, there is a maximum
point depending on pH and T, as can be seen in the graphic
Table 1 behavior of this polynomial Fig. 1a. For P (Fig. 1b), there is only a
Relative concentration of amino acids.
maximum point in terms of T, which is not as pronounced as DH.
Aminoacid residues/1000 residues This is because the linear term (P = 0.0002) is much more
Scales Hydrolysate significant than the quadratic one (P = 0.0349), which indicates
that the maximum point is very close to the upper limit of the
ASP 55.52 33.28
GLU 105.94 56.71 working range.
ASN 2.70 3.95
SER 46.10 32.44 DH ¼ 259:8 þ 31:0 pH þ 5:4 T  0:1 pHT  1:6 pH2  0:04 T 2
HIS 26.17 15.35 ð9Þ
GLY 346.89 610.29
TRE 15.48 5.81
CIT 9.91 18.89
ARG 0.88 2.36 P ¼ 9:9  1:1 pH þ 0:8 T  5:97x103  T 2 ð10Þ
ALA 11.08 3.86
TYR 47.76 90.35
CYS 13.96 6.58
VAL 9.68 18.20 Table 2
MET 6.89 5.86 P-values of the pH and temperature models for Alcalase1 2.4 L and Flavourzyme1
PHE 4.22 4.06 500 L.
ISO 15.60 13.09
LEU 4.97 4.77 Factor Alcalase1 2.4 L Flavourzyme1 500 L
LYS 12.03 10.04 DH (%) P (g/L) DH (%) P (g/L)
PRO 163.54 38.08
HYP 100.67 26.02 Model < 0.0001 0.0002 < 0.0001 0.0017
HAA 225.73 90.44 pH < 0.0001 0.0009 0.0002 –
AAA 51.98 94.42 T 0.2689 0.0002 < 0.0001 0.0075
EAA 95.94 79.54 pH*T 0.006 – – –
NEAA 904.06 920.46 pH2 0.0001 – – –
T2 < 0.0001 0.0349 – 0.0029
HAA: hydrophobic amino acids. Lack of fit 0.5752 0.0824 0.3793 0.5153
AAA aromatic amino acids. R2 0.9904 0.877 0.8955 0.7578
EAA essential amino acids. Adjusted R2 0.9836 0.836 0.8746 0.7039
NEAA non-essential amino acids.

4
L.M. Sierra-Lopera and J.E. Zapata-Montoya Biotechnology Reports 30 (2021) e00611

Fig. 1. Response surface for DH (a) and P (b) for Alcalase1 2.4 L.

3.4.1.2. Effects of pH and temperature on Flavourzyme1 500L. Eqs. 3.4.1.3. Optimization of the models to maximize DH and P. The
11 and 12 describe the behavior of DH and P using Flavourzyme1 models of Eqs. (9–12) were optimized with the aim of maximizing
500 L, respectively. The graphic behavior of Eqs. 11 and 12 is shown DH and P. The results obtained for optimal conditions of pH and T
in Fig. 2a and b, respectively. These models show that DH depends for Alcalase1 2.4 L are 8.1 and 58.5  C. For Flavourzyme1 500 L,
on both factors linearly, while P only depends on T in quadratic these values were found to be 8.3 and 53.8  C. It is worth noting the
terms. DH increases with pH and decreases with T in the working proximity of the experimental responses with the predicted ones,
range. However, P is independent of pH and is function of T. which corroborates the validity of the models and the power of the
response surface methodology as an optimization method.
DH ¼ 11:6  0:1 T þ 2:9 pH ð11Þ
However, the adjustment between predicted and experimental
values is better in the case of Alcalase1 2.4 L than for
Flavourzyme1 500 L. The predicted-experimental values for P
P ¼ 8:6 þ 0:5 T  4:041x103 T 2   ð12Þ were 6.9 g/L-6.7 g/L and 5.3 g/L-3.6 g/L for Alcalase1 2.4 L and
In general, Alcalase 1
2.4 L is more affected by pH than Flavourzyme1 500 L, respectively. On the other hand, predicted-
Flavourzyme1 500 L. In the first, pH increases in disfavor of DH experimental values for DH were 17.6 %-16.7 % and 6.4 %-5.9 % for
and P, whereas for Flavourzyme1 500 L, pH increases in favor of Alcalase1 2.4 L and Flavourzyme1 500 L, respectively.
DH. Other studies have found that increases in pH favor the yields Higher DH and P were obtained using Alcalase1 , which agrees
of protein liberation in tilapia skeletons [40]. However, very high with Gajanan et al., [46], because when the DH is higher the yield is
pH values can denature proteins and damage some amino acids also increased. Herman-lara et al., [47], found similar results in
[41]. hydrolysis processes with skeleton flour made from nile tilapia
The effects of pH are due to changes in its level, which affects (Oreochromis niloticus), indicating that DH is higher for Alcalase1
both substrate and enzyme. This is because it changes the 2.4 L since it is an endopeptidase, while Flavourzyme1 500 L is
distribution of charges and the conformation of proteins [42]. exopeptidase. These higher values for Alcalase1 2.4 L could be
Additionally, pH can influence the dissociation of active groups of because alkaline proteases show greater activity than neutral
the enzyme, affecting the dynamics of association of the enzyme proteases such as Flavourzyme1 500 L when acting on fish
with the substrate [43]. proteins [48–51]. The higher degree of hydrolysis in these
The effect of temperature on both enzymes can be explained by substrates is important because it may be associated with the
the fact that temperature increases, generally increase the kinetic increase in the levels of essential amino acids, hence the
energy of molecules. However, due to the protein nature of the nutritional value of the hydrolyzed products obtained [52].
enzymes, their tertiary structure is compromised when certain Generally, high DH in hydrolysates from fishery by-products have
temperature values are reached, which decreases the possibilities been linked with greater reducing capacity and radical capture
of union between the enzyme and the substrate by thermal [1,53]. The low molecular weights of peptides have been widely
denaturation of the enzyme [44]. This, in turn, leads to the loss of associated to the presence of biological activities [23]. The most
catalytic activity when working above a certain temperature limit interesting applications in this regard have been found in peptides
[45]. with molecular weights between 14 kDa [54]. Therefore,

Fig. 2. Response surface for DH (a) and P (b) for Flavourzyme1 500 L.

5
L.M. Sierra-Lopera and J.E. Zapata-Montoya Biotechnology Reports 30 (2021) e00611

obtaining hydrolysates with high degrees of hydrolysis (DH) seems concentrations in which substrate inhibition will not occur was
to increase the possibility of obtaining bioactive peptides [55]. defined at values between 0.5Km and 2.5Km [60].
Under the optimal conditions (temperature and pH) obtained
1 1
for each enzyme, the antioxidant activities were evaluated by ABTS V max ¼ ¼ ¼ 11:3 mM=s
Intercept 0:0884
(radical scavenging), FRAP (reducing power) and ICA (metal
chelating). The productivities in each bioactivity for Alcalase1
2.4 L were ABTS (223.1  4.9 mmol TE/g prot*mL enzyme), FRAP
Km ¼ V max slope ¼ 11:316:6 ¼ 188:2 mM ¼ 21:8 g=L
(13.1  0.7 mmol TE/g prot*mL enzyme) and ICA (61  1%). Mean-
while for Flavourzyme1 500 L ABTS (53.7  0.6 mmol TE/g prot*mL The high Km values indicate that the enzyme requires high
enzyme), FRAP (6.3  0.4 mmol TE/g prot*mL enzyme) and ICA substrate concentrations to reach saturation and achieve maxi-
(65  3%). mum speed, this parameter is associated with the affinity between
The comparation according to the type of enzyme show that the the enzyme and the substrate [60]. In this sense, the Km value
protease affects the functionality of the hydrolysates obtained [56]. obtained for the substrate is greater than the obtained for
In this case, the difference is due to the fact that Alcalase1 2.4 L is Alcalase1 2.4 L in other substrates such as tilapia viscera (1.9 g/
an endopeptidase, which breaks peptide bonds from non-terminal L) [61] and Salmon muscle (4 g/L) [30]. This is possibly due to the
amino acids, while Flavourzyme1 500 L is an exopeptidase that difference in the scale's structure, which has a higher mineral
breaks the N-termini of chains [5]. Alcalase1 2.4 L and Fla- content, partially comprised of hydroxyapatite (Ca10(PO4)6(OH)2
vourzyme1 500 L liberates higher residues with hydrophobic [62], which makes it a protease resistant structure [63].
amino acids, which have been associated with antioxidant In the second experimental design, the DH and P are still shown
activities. Flavourzyme1 500 L liberates free amino acids [1,57], to corroborate the assumption of the relationship between DH and
and amino acids such as proline have been associated with the antioxidant activity. The effect of enzyme (E) and substrate (S)
increase in antioxidant potential thanks to the ability to donate concentrations on P, DH and the biological activities ABTS and ICA,
protons [58]. Other amino acids with acidic residues such as was evaluated by means of a central composite design. The
glutamic acid are important contributors to the antioxidant respective analysis of variance is shown in Table 3.
activity of hydrolysates [59]. Table 3 shows that all of the response models were significant,
These results agree with those reported by Karama
c, 2016 [56] without a significant lack of fit adjustment. However, for the P and
who found that Alcalase1 2.4 L is superior to Flavourzyme1 500 L ICA models, only the substrate concentration was significant, while
in terms of antioxidant activity. for DH, the linear effects of both factors were significant, as well as
Given its higher yields of protein liberation and productivity of the quadratic term of the substrate concentration. Finally, ABTS
hydrolysates with antioxidant capacity, these results suggest that was explained by the linear and quadratic effects of the two
Alcalase1 2.4 L is the enzyme with the greatest potential to obtain variables, with second order interactions.
antioxidant peptides from red tilapia scales. The models were obtained for each of the responses from the
To fully define the operating conditions of the enzymatic ANOVA (Table 3) (Eqs. 13–16) once the non-significant terms were
hydrolysis, it was necessary to define the enzyme and substrate eliminated. This was done without affecting the hierarchy of the
concentrations in which the enzyme could perform better. In this design and the transformed variable Y = DH2.55 was used instead of
sense, it was convenient to determine the kinetic parameters of DH.
Alcalase1 2.4 L with this particular substrate, by means of a
saturation curve, in which the initial velocity vs. the substrate ðDHÞ2:55 ¼ 1163:1  204:9 S þ 488:3E  231:9 E2   ð13Þ
concentration is plotted [60] as shown in Fig. 3. This figure shows
that above 200 mM, increases in substrate concentration do not
generate an increase in the speed of the reaction, possibly because P ¼ 25:6 þ 11:2 S  ð14Þ
there is a saturation of the enzyme caused by the substrate.
The kinetic parameters were determined using the Lineweaver-
Burk model, by graphing the multiplicative inverses of the ABTS ¼ 230:8 þ 2:1 S þ 68:7 E  35:0 ES þ 33:0 S2  43:9 E2
mentioned variables [60]. After adjusting a linear model to this ð15Þ
curve, the slope values and the intercept can be obtained. Then,
using these values, Vmax was calculated: 11.3 mM/sec and Km:
188.2 mM or 21.8 g/L. Based on this data, the range of substrate ICA ¼ 82:3  4:2 S   ð16Þ
Fig. 4 shows the response surfaces of the models found for each
of the response variables. Those results indicate that S helps to
increase P but decrease ICA while E improves ABTS and DH. P and S
increase simultaneously because the variables are directly related.

Table 3
ANOVA of models and variables chosen on P, DH, ABTS and ICA.

Variable P (g/L) DH (%) ABTS ICA

Model < 0.0001 < 0.0001 0.0013 0.0094
S < 0.0001 0.0072 0.8243 0.0094
E – < 0.0001 0.0002 –
S2 – 0.0103 0.0132 –
E2 – – 0.0060 –
S*E – – 0.0332 –
Lack of fit 0.3047 0.1549 0.0730 0.7405
R2 0.9787 0.8937 0.9147 0.5065
Adjusted R2 0.9766 0.8582 0.8538 0.4571

Fig. 3. Substrate saturation curve of Alcalase1 2.4 L.

6
L.M. Sierra-Lopera and J.E. Zapata-Montoya Biotechnology Reports 30 (2021) e00611

Fig. 4. Response surfaces for DH (a) and P (b), ABTS (c) and ICA (d) at different levels of S and E.

On the other hand, having a higher E in the system also favors the reported by Blanco et al. [67] in the hydrolysis of skin collagen of
probability of the enzyme-substrate encounter and therefore the various species using Alcalase1 , which was between 12 % and 16 %.
DH. If E increases ABTS could decrease because the enzyme would Likewise, the antioxidant activity obtained is higher than that
be hydrolyzing peptides. S has opposite effects on ABTS depending reported by some authors such as Sai-Ut, Benjakul, Sumpavapol, &
on the zone of E in which it is located. At low values of E, S has a Kishimura [68], who found that the antioxidant activity by ABTS
positive effect on ABTS, but at high values of E, it has negative was 80 mmol TE/g sample for gelatin hydrolysates from Aluterus
effects. These differences in ABTS as a function of S, are due to the monoceros. In this study a 190 mmol TE/g sample was obtained.
fact that the affinity of the enzyme towards the protein is Respecting to ICA, the results obtained are similar to those
determined by the balance between the native protein and the one reported by Choonpicharn & Jaturasitha [69], who found 77 %
deployed in the solution [64]. Changes in this balance may suggest activity in hydrolysates of skin gelatin from nile tilapia.
changes in hydrolysis mechanisms. It has been shown that the The hydrolysis process was carried out in optimal conditions of
affinity for the intact protein was different in the hydrolysis with pH, T, S and E. Data was collected from each of the response
Alcalase1 2.4 L when using different substrate concentrations. variables as a function of time and DH to assess the moment and
That is, that the composition of these hydrolysates was different DH that generated the greatest biological activities.
and that, in effect, this is caused by changes in the selectivity of the Table 5 shows the values of the responses as a function of time
enzyme in varied operating conditions [65]. and DH using Alcalase1 2.4 L. The study found that at the 10-
The models presented in Eqs. 5–8 were optimized to maximize minute point (DH of 7%) most of the protein had solubilized.
DH, P and biological activities (ABTS and ICA). The study found that However, by that time the biological activities were very low, so it
the substrate concentration should be at 45 g/L and the enzyme was necessary to continue hydrolysis for a longer time (517 min), to
concentration at 4.42 g/L. The response variables obtained were reach a degree of hydrolysis of 18 %.
verified experimentally (Table 4). When comparing the evolution of the responses between
Protein liberation had a yield of 93 %, which was expected since 55 min (DH 12 %) and 517 min (DH 18 %), the analysis found that the
Alcalase1 2.4 L allows the solubilization of fish proteins and is also increases in biological activity did not correspond to the time
highly efficient in the hydrolysis of this type of materials [66]. On interval (462 min) and P had remained constant. After this
the other hand, the degree of hydrolysis obtained is close to that discovery, the hydrolysis time was kept at 55 min, which was
sufficient to bring the reaction to the limit productivity level.
Table 4 The results shown in Table 5 indicate that the reaction followed
Predicted and experimental values for P, DH, ABTS and ICA under optimal conditions
a typical behavior, with a high exchange rate at first, followed by a
of S and E.
slowdown until reaching a quasi-stationary state in the last
Variable Predicted Experimental Absolute error minutes [70]. This behavior has been attributed to one or several of
P (g/L) 36.78 41.9  1.19 5.1 these three factors: (a) a decrease in the concentration of the
DH (%) 15.85 15.65  0.33 0.2 peptide bonds that are susceptible to be hydrolysed by proteases
ABTS (mmol TE/g prot) 271.15 209.8  2.96 61.3 [71], (b) possible inhibition of enzymes caused by the hydrolysis
ICA (%) 78 77  1.41 1.0
substrate [61], (c) thermal denaturation of the enzyme [72].

7
L.M. Sierra-Lopera and J.E. Zapata-Montoya Biotechnology Reports 30 (2021) e00611

Table 5
Behavior of the response variables as a function of time and DH.

Time (min) DH (%) P (g/L) ABTS (mmol TE/g prot) ORAC (mmol TE/g prot) ICA (%)
1.2 2.0 37.2  1.27 115.6  1.07 129.5  2.00 15.8  1.30
10.0 7.0 41.4  1.99 200.5  11.7 176.6  11.78 54.9  3.41
55.0 12.0 41.9  1.19 257.5  7.00 195.0  8.78 70.7  3.42
180.0 15.6 42.1  0.87 209.8  2.96 156.5  6.77 77  1.41
517.0 18.0 43.8  1.19 289.6  1.05 141.2  18.78 85.4  2.38

Table 6
DH, P, ABTS, FRAP AND ICA at different stirring rates.

Stirring rate (rpm) DH (%) P (g/L) ABTS (mmol TE/g prot) FRAP (mmol TE/g prot) ICA (%)
250 9.9 30.3  1.3 245.3  10.7 38.7  2.5 33.4  5.1
350 12.0 37.4  1.5 199.8  8.5 40.3  3.1 66.2  4.8
Reactor 0.5 L 12.0 41.9  1.2 257.5  7.0 28.4  5.6 70.7  3.4

3.5. Scale up the optimal conditions of the enzymatic hydrolysis CRediT authorship contribution statement

During the scaling up process, it is important to prove that Leidy Maritza Sierra-Lopera: Performed the experiments;
the same results obtained on a small scale can be achieved on Analyzed and interpreted the data; Wrote the paper. José Edgar
the larger scale in terms of efficiency and quality. However, the Zapata-Montoya: Conceived and designed the experiments;
scaling of fish protein hydrolysis has not been studied Analyzed and interpreted the data; Contributed reagents, materi-
sufficiently [72]. When the process was scaled to 7.5 L, the als, analysis tools or data; Wrote the paper.
analysis found that the biological activities differed from those
obtained in the 0.5 L reactor. In this vein, the agitation speed was Declaration of Competing Interest
modified to determine the effect of this variable on the
responses. Table 6 shows the results of the variables evaluated The authors declare that they have no known competing
for the 7.5 L reactor, with two different stirring speeds (250 and financial interests or personal relationships that could have
350 rpm), in contrast to the results obtained in the 0.5 L reactor. appeared to influence the work reported in this paper.
Table 6 shows that ABTS and ICA decrease when the speed of
agitation is 250 rpm, however, FRAP increases. These results Acknowledgements
indicate that it is possible to obtain a hydrolysate with
antioxidant activity by increasing the volume of work, as found The authors thank the Antioquia Department through “Sistema
by other authors in processes of hydrolysis of various materials General de Regalías de Colombia” and Comité para el Desarrollo de
such as cauliflower [73], sucrose [74] and lignocellulosic la Investigación en la Universidad de Antioquia (CODI) through
substrates [75]. sustainability program and Colciencias for financial support.
It is important to note that the agitation speed favors DH and P,
because it improves the homogeneity of the system, but the effect References
on the activities of the peptides obtained underlies the conforma-
tional changes that the enzyme undergoes when subjected to [1] L.J. Gómez, N.A. Gómez, J.E. Zapata, G. López-García, A. Cilla, A. Alegría, In-vitro
antioxidant capacity and cytoprotective/cytotoxic effects upon Caco-2 cells of
different flow conditions [76]. red tilapia (Oreochromis spp.) viscera hydrolysates, Food Res. Int. 120 (2019)
52–61, doi:http://dx.doi.org/10.1016/j.foodres.2019.02.029.
4. Conclusions [2] O. Martínez-alvarez, S. Chamorro, A. Brenes, Protein hydrolysates from animal
processing by-products as a source of bioactive molecules with interest in
animal feeding : a review, Food Res. Int. 73 (2015) 204–212.
The amount of protein released and the degree of hydrolysis in [3] O. Villamil, H. Váquiro, J.F. Solanilla, Fish viscera protein hydrolysates:
the enzymatic hydrolysis of red Tilapia (Oreochromis sp) scales, is production, potential applications and functional and bioactive properties,
Food Chem. 224 (2017) 160–171, doi:http://dx.doi.org/10.1016/j.
significantly affected by the pH and the temperature of the foodchem.2016.12.057.
medium. This phenomenon holds true when the material is [4] C. Huang, J. Kuo, S. Wu, H. Tsai, Isolation and characterization of fish scale
hydrolyzed with Alcalase1 2.4 L or Flavourzyme1 500 L. The collagen from tilapia (Oreochromis sp.) by a novel extrusion – hydro-
productivity of the hydrolysates expressed as mmol TE/gprot* mL extraction process, Food Chem. 190 (2016) 997–1006, doi:http://dx.doi.org/
10.1016/j.foodchem.2015.06.066.
enzyme, is higher for Alcalase1 2.4 L than for Flavourzyme1 500 L. [5] P. Ambigaipalan, A.S. Al-khalifa, Antioxidant and angiotensin I converting
The kinetics of Alcalase1 2.4 L in this reaction can be described by enzyme (ACE) inhibitory activities of date seed protein hydrolysates prepared
means of the Michaelis-Menten model, with a Linewaver-Burk using Alcalase, Flavourzyme and Thermolysin, J. Funct. Foods 18 (2015) 1125–
1137, doi:http://dx.doi.org/10.1016/j.jff.2015.01.021.
linearization. The antioxidant activity of the hydrolysates obtained [6] L. Najafian, A.S. Babji, Production of bioactive peptides using enzymatic
with Alcalase1 2.4 L depend on substrate and enzyme concentra- hydrolysis and identification antioxidative peptides from patin (Pangasius
tion. The iron chelating activity only depends on the concentration sutchi) sarcoplasmic protein hydolysate, J. Funct. Foods 9 (2014) 280–289, doi:
http://dx.doi.org/10.1016/j.jff.2014.05.003.
of the substrate. The optimal conditions to obtain antioxidant [7] Pa. Harnedy, R.J. FitzGerald, Bioactive peptides from marine processing waste
peptides from red tilapia scale are pH 8.1, T 58.5  C, S 45 g/L and E and shellfish: A review, J. Funct. Foods 4 (2012) 6–24, doi:http://dx.doi.org/
4.42 g/L. The scaling of the reaction to a 7.5 L reactor using the 10.1016/j.jff.2011.09.001.
[8] S. Benjakul, S. Yarnpakdee, T. Senphan, S.M. Halldorsdottir, H.G. Kristinsson,
dynamic similarity criterion based on the Reynolds number, allows Fish Protein Hydrolysates: Production, Bioactivities, and Applications, (2014) .
for the reproduction of the conditions obtained in the 0.5 L reactor. [9] X. Wang, H. Yu, R. Xing, P. Li, Characterization, Preparation, and Purification of
However, the reaction is affected by the impeller speed in the 7.5 L Marine Bioactive Peptides, Biomed Res. Int. (2017), doi:http://dx.doi.org/
10.1155/2017/9746720 2017.
reactor.

8
L.M. Sierra-Lopera and J.E. Zapata-Montoya Biotechnology Reports 30 (2021) e00611

[10] W. Weng, L. Tang, B. Wang, J. Chen, W. Su, K. Osako, M. Tanaka, Antioxidant [36] R. Bansil, B.S. Turner, Mucin structure, aggregation, physiological functions and
properties of fractions isolated from blue shark (Prionace glauca) skin gelatin biomedical applications, Curr. Opin. Colloid Interface Sci. 11 (2006) 164–170.
hydrolysates, J. Funct. Foods 11 (2014) 342–351, doi:http://dx.doi.org/10.1016/ [37] A.T. Girgih, R. He, S. Malomo, M. Offengenden, J. Wu, R.E. Aluko, Structural and
j.jff.2014.10.021. functional characterization of hemp seed (Cannabis sativa L.) protein-derived
[11] R. Nasri, I. Ben, A. Bougatef, N. Nedjar-arroume, P. Dhulster, J. Gargouri, M. antioxidant and antihypertensive peptides, J. Funct. Foods 6 (2014) 384–394.
Karra, M. Nasri, Anticoagulant activities of goby muscle protein hydrolysates, [38] J.Y. Je, S.Y. Park, J.Y. Hwang, C.B. Ahn, Amino acid composition and in vitro
Food Chem. 133 (2012) 835–841, doi:http://dx.doi.org/10.1016/j. antioxidant and cytoprotective activity of abalone viscera hydrolysate, J. Funct.
foodchem.2012.01.101. Foods 16 (2015) 94–103, doi:http://dx.doi.org/10.1016/j.jff.2015.04.023.
[12] S. Choonpicharn, S. Jaturasitha, N. Rakariyatham, N. Suree, H. Niamsup, [39] T. Aspevik, H. Egede-Nissen, Å. Oterhals, A systematic approach to the
Antioxidant and antihypertensive activity of gelatin hydrolysate from Nile comparison of cost efficiency of endopeptidases for the hydrolysis of Atlantic
tilapia skin, J. Food Sci. Technol. 52 (2015) 3134–3139. salmon (Salmo salar) by-products, Food Technol. Biotechnol. 54 (2016) 421–
[13] N. Charoenphun, B. Cheirsilp, N. Sirinupong, W. Youravong, Calcium-binding 431, doi:http://dx.doi.org/10.17113/ftb.54.04.16.4553.
peptides derived from tilapia (Oreochromis niloticus) protein hydrolysate, Eur. [40] C. Chomnawang, J. Yongsawatdigul, Journal of aquatic food product protein
Food Res. Technol. 236 (2013) 57–63, doi:http://dx.doi.org/10.1007/s00217- recovery of Tilapia frame by- products by pH-Shift method, J. Aquat. Food Prod.
012-1860-2. Technol. 22 (2013) 112–119, doi:http://dx.doi.org/10.1080/
[14] P. Chuesiang, R. Sanguandeekul, Protein hydrolysate from tilapia frame: 10498850.2011.629077.
antioxidant and angiotensin I converting enzyme inhibitor properties, Int. J. [41] A.K. Anal, A. Noomhorm, P. Vongsawasdi, Protein hydrolysates and bioactive
Food Sci. Technol. 50 (2015) 1436–1444, doi:http://dx.doi.org/10.1111/ peptides from seafood and crustacean waste: their extraction, bioactive
ijfs.12762. properties and industrial perspectives, Mar. Proteins Pept. Biol. Act. Appl.
[15] D. Chen, X. Mu, H. Huang, R. Nie, Z. Liu, M. Zeng, Isolation of a calcium-binding (2013) 709–735, doi:http://dx.doi.org/10.1002/9781118375082 ch36.
peptide from tilapia scale protein hydrolysate and its calcium bioavailability in [42] J. Roslan, K.F.M. Yunos, N. Abdullah, S.M.M. Kamal, Characterization of fish
rats, J. Funct. Foods 6 (2014) 575–584, doi:http://dx.doi.org/10.1016/j. protein hydrolysate from Tilapia (Oreochromis niloticus) by-product, Agric.
jff.2013.12.001. Agric. Sci. Procedia 2 (2014) 312–319, doi:http://dx.doi.org/10.1016/j.
[16] J. Xia, G. Wang, J. Lin, Y. Wang, J. Chu, Advances and practices of bioprocess, aaspro.2014.11.044.
Adv. Biochem. Eng. Biotechnoldvance Biochem. Eng. Biotechnol. (2015), doi: [43] J.A. Morales, O.A. Figueroa, J.E. Zapata, Optimización de hidrólisis enzimática
http://dx.doi.org/10.1007/10. de la fracción globular de sangre bovina por metodología de superficie
[17] G.W. Latimer, Official Methods of Analysis of AOAC International, AOAC respuesta y evaluación de sus propiedades antioxidantes, Inf. Tecnol. 28 (2017)
International, Gaithersburg, Md, 2012 ISBN: 978-0-935584-83-7. 75–86, doi:http://dx.doi.org/10.4067/S0718-07642017000200009.
[18] J.W. Henderson, A. Brooks, Improved amino acid methods using agilent [44] A. Noman, Y. Xu, W.Q. AL-Bukhaiti, S.M. Abed, A.H. Ali, A.H. Ramadhan, W. Xia,
ZORBAX eclipse plus C18 columns for a variety of agilent LC instrumentation Influence of enzymatic hydrolysis conditions on the degree of hydrolysis and
and separation goals, Agil. Technol. (2010) 1–16. functional properties of protein hydrolysate obtained from Chinese sturgeon
[19] J. Adler-Nissen, Determination of the degree of hydrolysis of food protein (Acipenser sinensis) by using papain enzyme, Process Biochem. 67 (2018) 19–
hydrolysates by trinitrobenzenesulfonic acid, J. Agric. Food Chem. 27 (1979) 28.
1256–1262. [45] R. Pérez-Gálvez, M.C. Almécija, F.J. Espejo, E.M. Guadix, A. Guadix, Bi-objective
[20] S. Khueychai, N. Jangpromma, K. Choowongkomon, A. Joompang, S. Daduang, optimisation of the enzymatic hydrolysis of porcine blood protein, Biochem.
M. Vesaratchavest, W. Payoungkiattikun, S. Tachibana, S. Klaynongsruang, A Eng. J. 53 (2011) 305–310.
novel ACE inhibitory peptide derived from alkaline hydrolysis of ostrich [46] P.G. Gajanan, K. Elavarasan, B.A. Shamasundar, Bioactive and functional
(Struthio camelus) egg white ovalbumin, Process Biochem. 73 (2018) 235–245, properties of protein hydrolysates from fish frame processing waste using
doi:http://dx.doi.org/10.1016/j.procbio.2018.07.014. plant proteases, Environ. Sci. Pollut. Res. 23 (2016) 24901–24911, doi:http://dx.
[21] J. Adler-Nissen, Enzymic Hydrolysis of Food Proteins, Elsevier applied science doi.org/10.1007/s11356-016-7618-9.
publishers, 1986. [48] J. Dumay, C. Donnay-Moreno, G. Barnathan, P. Jaouen, J.-P. Bergé, Improvement
[22] P. Valencia, K. Espinoza, A. Ceballos, M. Pinto, S. Almonacid, Novel modeling of lipid and phospholipid recoveries from sardine (Sardina pilchardus) viscera
methodology for the characterization of enzymatic hydrolysis of proteins, using industrial proteases, Process Biochem. 41 (2006) 2327–2332.
Process Biochem. 50 (2015) 589–597, doi:http://dx.doi.org/10.1016/j. [49] V. Klompong, S. Benjakul, D. Kantachote, K.D. Hayes, F. Shahidi, Comparative
procbio.2014.12.028. study on antioxidative activity of yellow stripe trevally protein hydrolysate
[23] S. Zhong, S. Liu, J. Cao, S. Chen, W. Wang, X. Qin, Fish protein isolates recovered produced from Alcalase and Flavourzyme, Int. J. Food Sci. Technol. 43 (2008)
from silver carp (Hypophthalmichthys molitrix) by-products using alkaline pH 1019–1026, doi:http://dx.doi.org/10.1111/j.1365-2621.2007.01555.x.
solubilization and precipitation, J. Aquat. Food Prod. Technol. 25 (2016) 400– [50] S. Yarnpakdee, S. Benjakul, H.G. Kristinsson, H. Kishimura, Antioxidant and
413, doi:http://dx.doi.org/10.1080/10498850.2013.865282. sensory properties of protein hydrolysate derived from Nile tilapia
[24] S.K.C. Chang, Y. Zhang, Protein Analysis BT - Food Analysis, (2017), doi:http:// (Oreochromis niloticus) by one- and two-step hydrolysis, J. Food Sci. Technol.
dx.doi.org/10.1007/978-3-319-45776-5_18. 52 (2015) 3336–3349, doi:http://dx.doi.org/10.1007/s13197-014-1394-7.
[25] L. Zheng, M. Zhao, C. Xiao, Q. Zhao, G. Su, Practical problems when using ABTS [51] J. Gunasekaran, N. Kannuchamy, S. Kannaiyan, R. Chakraborti, V. Gudipati,
assay to assess the radical-scavenging activity of peptides : importance of Protein hydrolysates from shrimp (Metapenaeus dobsoni) head waste:
controlling reaction pH and time, Food Chem. 192 (2016) 288–294, doi:http:// optimization of extraction conditions by response surface methodology, J.
dx.doi.org/10.1016/j.foodchem.2015.07.015. Aquat. Food Prod. Technol. 24 (2015) 429–442.
[26] R. Pulido, L. Bravo, F. Saura-Calixto, Antioxidant activity of dietary polyphenols [52] S. Yarnpakdee, S. Benjakul, H.G. Kristinsson, H. Kishimura, Antioxidant and
as determined by a modified ferric reducing/antioxidant power assay, J. Agric. sensory properties of protein hydrolysate derived from Nile tilapia
Food Chem. 48 (2000) 3396–3402. (Oreochromis niloticus) by one-and two-step hydrolysis, J. Food Sci. Technol.
[27] J.P. Adjimani, P. Asare, Antioxidant and free radical scavenging activity of iron 52 (2015) 3336–3349.
chelators, Toxicol. Reports. 2 (2015) 721–728, doi:http://dx.doi.org/10.1016/j. [53] P.J. García-Moreno, I. Batista, C. Pires, N.M. Bandarra, F.J. Espejo-Carpio, A.
toxrep.2015.04.005. Guadix, E.M. Guadix, Antioxidant activity of protein hydrolysates obtained
[28] S.A. Hamid, N.R.A. Halim, N.M. Sarbon, Optimization of enzymatic hydrolysis from discarded Mediterranean fish species, Food Res. Int. 65 (2014) 469–476,
conditions of golden apple snail (Pomacea canaliculata) protein by Alcalase, doi:http://dx.doi.org/10.1016/j.foodres.2014.03.061.
Int. Food Res. J. 22 (2015) 1615. [54] M. Opheim, R. Šližyte, _ H. Sterten, F. Provan, E. Larssen, N.P. Kjos, Hydrolysis of
[30] P. Valencia, M. Pinto, S. Almonacid, Identification of the key mechanisms Atlantic salmon (Salmo salar) rest raw materials—effect of raw material and
involved in the hydrolysis of fish protein by Alcalase, Process Biochem. 49 processing on composition, nutritional value, and potential bioactive peptides
(2014) 258–264, doi:http://dx.doi.org/10.1016/j.procbio.2013.11.012. in the hydrolysates, Process Biochem. 50 (2015) 1247–1257.
[31] B. Palmqvist, A. Kadi
c, K. Hägglund, A. Petersson, G. Lidén, Scale-up of high- [55] S. Beaubier, X. Framboisier, I. Ioannou, O. Galet, R. Kapel, Simultaneous
solid enzymatic hydrolysis of steam-pretreated softwood: the effects of quantification of the degree of hydrolysis, protein conversion rate and mean
reactor flow conditions, Biomass Convers. Biorefinery. 6 (2016) 173–180. molar weight of peptides released in the course of enzymatic proteolysis, J.
[32] K. Matmaroh, S. Benjakul, T. Prodpran, A.B. Encarnacion, H. Kishimura, Chromatogr. B Anal. Technol. Biomed. Life Sci. 1105 (2019) 1–9, doi:http://dx.
Characteristics of acid soluble collagen and pepsin soluble collagen from scale doi.org/10.1016/j.jchromb.2018.12.005.
of spotted golden goatfish (Parupeneus heptacanthus), Food Chem. 129 (2011) [56] M. Karama
c, A. Kosin
ska-Cagnazzo, A. Kulczyk, Use of different proteases to
1179–1186, doi:http://dx.doi.org/10.1016/j.foodchem.2011.05.099. obtain flaxseed protein hydrolysates with antioxidant activity, Int. J. Mol. Sci.
[33] A.A. El-Rashidy, A. Gad, A.E.H.G. Abu-Hussein, S.I. Habib, N.A. Badr, A.A. 17 (2016), doi:http://dx.doi.org/10.3390/ijms17071027.
Hashem, Chemical and biological evaluation of Egyptian Nile Tilapia [57] A.T. Girgih, R. He, F.M. Hasan, C.C. Udenigwe, T.A. Gill, R.E. Aluko, Evaluation of
(Oreochromis niloticas) fish scale collagen, Int. J. Biol. Macromol. 79 (2015) the in vitro antioxidant properties of a cod (Gadus morhua) protein
618–626, doi:http://dx.doi.org/10.1016/j.ijbiomac.2015.05.019. hydrolysate and peptide fractions, Food Chem. 173 (2015) 652–659, doi:http://
[34] S. Wangtueai, A. Noomhorm, LWT - food science and technology processing dx.doi.org/10.1016/j.foodchem.2014.10.079.
optimization and characterization of gelatin from lizardfish (Saurida spp.) [58] M. Silveira Coelho, S. de Araujo Aquino, J. Machado Latorres, M. De las
scales, LWT - Food Sci. Technol. 42 (2009) 825–834, doi:http://dx.doi.org/ Mercedes Salas-Mellado, in vitro and in vivo antioxidant capacity of chia
10.1016/j.lwt.2008.11.014. protein hydrolysates and peptides, Food Hydrocoll. 91 (2019) 19–25, doi:
[35] M. Gauza-Włodarczyk, L. Kubisz, D. Włodarczyk, Amino acid composition in http://dx.doi.org/10.1016/j.foodhyd.2019.01.018.
determination of collagen origin and assessment of physical factors effects, Int. [59] C.C. Udenigwe, R.E. Aluko, Chemometric Analysis of the Amino Acid
J. Biol. Macromol. 104 (2017) 987–991, doi:http://dx.doi.org/10.1016/j. Requirements of Antioxidant Food Protein Hydrolysates, (2011), pp. 3148–
ijbiomac.2017.07.013. 3161, doi:http://dx.doi.org/10.3390/ijms12053148.

9
L.M. Sierra-Lopera and J.E. Zapata-Montoya Biotechnology Reports 30 (2021) e00611

[60] R. a Copeland, a J. Wiley, ENZYMES A Practical Introduction, (2000), doi:http:// produced by protease from B. Amyloliquefaciens H11, Dry. Technol. 32 (2014)
dx.doi.org/10.1021/jm020467f. 1552–1559.
[61] J.E. Zapata Montoya, D.E. Giraldo-Rios, A.J. Baéz-Suarez, Kinetic modeling of [69] S. Choonpicharn, S. Jaturasitha, Antioxidant and Antihypertensive Activity of
the enzymatic hydrolysis of proteins of visceras from red tilapia (Oreochromis Gelatin Hydrolysate from Nile Tilapia Skin, (2014), doi:http://dx.doi.org/
sp.): effect of substrate and enzyme concentration, Vitae 25 (2018) 17–25, doi: 10.1007/s13197-014-1581-6.
http://dx.doi.org/10.17533/udea.vitae.v25n1a03. [70] O.A. Figueroa, J.E. Zapata, C.P. Sánchez, Optimización de la hidrólisis
[62] M. Okuda, N. Ogawa, M. Takeguchi, A. Hashimoto, M. Tagaya, S. Chen, N. enzimática de proteínas de plasma bovino, Inf. Tecnol. 27 (2016) 39–52,
Hanagata, T. Ikoma, Minerals and aligned collagen fibrils in tilapia fish scales: doi:http://dx.doi.org/10.4067/S0718-07642016000200006.
structural analysis using dark-field and energy-filtered transmission electron [71] D. Kumar, M.K. Chatli, Enzymatic Hydrolysis of Camel Milk Casein and Its
microscopy and electron tomography, Microsc. Microanal. 17 (2011) 788–798. Antioxidant Properties, (2016), doi:http://dx.doi.org/10.1007/s13594-015-0275-9.
[63] Y. Zhang, D. Tu, Q. Shen, Z. Dai, Fish scale valorization by hydrothermal [72] J.E. Zapata, M. Moya, O.A. Figueroa, Hidrólisis Enzimática de la Proteína de
pretreatment followed by enzymatic hydrolysis for gelatin hydrolysate Vísceras de Trucha Arco íris (Oncorhynchus mykiss): efecto del tipo de enzima,
production, Molecules 24 (2019) 1–14, doi:http://dx.doi.org/10.3390/ Temperatura, pH y velocidad de agitación, Inf. Tecnológica 30 (2019) 63–72.
molecules24162998. [73] C.M. Montone, A.L. Capriotti, C. Cavaliere, G. La Barbera, S. Piovesana, R.Z.
[64] Y. Deng, C.I. Butré, P.A. Wierenga, Influence of substrate concentration on the Chiozzi, A. Laganà, Characterization of antioxidant and angiotensin-converting
extent of protein enzymatic hydrolysis, Int. Dairy J. 86 (2018) 39–48. enzyme inhibitory peptides derived from cauli fl ower by-products by
[65] C.I. Butre’, P.A. Wierenga, H. Gruppen, Effects of ionic strength on the multidimensional liquid chromatography and bioinformatics, J. Funct. Foods
enzymatic hydrolysis of diluted and concentrated whey protein isolate, J. 44 (2018) 40–47, doi:http://dx.doi.org/10.1016/j.jff.2018.02.022.
Agric. Food Chem. 60 (2012) 5644–5651. [74] D. Martínez, C. Menéndez, L. Hernández, A. Sobrino, L.E. Trujillo, I. Rodríguez, E.
[66] S.N.K.M. Putra, N.H. Ishak, N.M. Sarbon, Biocatalysis and Agricultural R. Pérez, Scaling-up batch conditions for ef fi cient sucrose hydrolysis catalyzed
Biotechnology Preparation and characterization of physicochemical by an immobilized recombinant Pichia pastoris cells in a stirrer tank reactor,
properties of golden apple snail (Pomacea canaliculata) protein hydrolysate as EJBT 25 (2017) 39–42, doi:http://dx.doi.org/10.1016/j.ejbt.2016.11.003.
a ff ected by di ff erent proteases, Biocatal. Agric. Biotechnol. 13 (2018) 123– [75] V. Sotaniemi, S. Taskila, H. Ojamo, J. Tanskanen, Biomass and Bioenergy
128, doi:http://dx.doi.org/10.1016/j.bcab.2017.12.002. Controlled feeding of lignocellulosic substrate enhances the performance of
[67] M. Blanco, J.A. Vázquez, R.I. Pérez-Martín, C.G. Sotelo, Hydrolysates of fish skin fed-batch enzymatic hydrolysis in a stirred tank reactor, Biomass Bioenergy 91
collagen: an opportunity for valorizing fish industry byproducts, Mar. Drugs 15 (2016) 271–277, doi:http://dx.doi.org/10.1016/j.biombioe.2016.05.037.
(2017) 131. [76] L.J. Gómez, J.E. Zapata, Efecto del nivel de grasa y velocidad de agitación en la
[68] S. Sai-Ut, S. Benjakul, P. Sumpavapol, H. Kishimura, Effect of drying methods on hidrolisis enzimática de vísceras de tilapia roja (orechromis sp.), Inf. Tecnol. 28
odorous compounds and antioxidative activity of gelatin hydrolysate (2017) 47–56, doi:http://dx.doi.org/10.4067/S0718-07642017000400007.

10