Received: 25 January 2019 Revised: 3 April 2019 Accepted: 18 April 2019
DOI: 10.1111/jfpe.13084
OR I G I N A L A R T I C L E
Antioxidant activities of sunflower protein hydrolysates
treated with dual-frequency ultrasonic: Optimization study
Mokhtar Dabbour1,2 | Ronghai He1 | Benjamin Mintah1,3 | Haile Ma1
1School of Food and Biological Engineering,
Jiangsu University, Zhenjiang, China Abstract
2Department of Agricultural and Biosystems In this study, ultrasound pretreatments of sunflower-meal protein (SMP) to yield high anti-
Engineering, Faculty of Agriculture, Benha
oxidant capacity from its hydrolysates by response surface methodology were optimized.
University, Moshtohor, Qaluobia, Egypt
3ILSI-UG FSNTC, Department of Nutrition and Optimization of experimental conditions was achieved to examine the impact of tempera-
Food Science, University of Ghana, Legon, ture, solvent-solid ratio and sonication time on antioxidant capacities of SMP hydrolysates
Accra, Ghana
with Box–Behnken's design. Quadratic models of DPPH-scavenging activity (DPPHSA),
Correspondence hydroxyl-radical scavenging activity (HRSA), and Cu2+ and Fe2+ chelating activity (Cu2+-CA
Ronghai He and Mokhtar Dabbour, School of
Food and Biological Engineering, Jiangsu and Fe
2+-CA) were developed, and their coefficients observed from multiple-regression
University, 301 Xuefu Road, Zhenjiang analysis. ANOVA indicated that time was highly significant (p < .01) on all experimental
212013, China.
Emails: heronghai1971@126.com (R. H.) and responses. The best experimental point of DPPHSA, HRSA, Cu
2+-CA, and Fe2+-CA was
mokhtar.dabbour@fagr.bu.edu.eg (M. D.) accessed at 42.50
C, 18.16 mL/g and 26.52 min and the predicted data for these
Funding information responses were 52.09, 70.05, 50.85, and 43.35%, respectively. Outcome of verification
National Primary Research and Development experiment was reliable with predicted data for all responses. Additionally, DPPHSA,
Plan of Jiangsu Province, Grant/Award 2+ 2+
Number: 2016YFD0401401; Primary Research HRSA, Cu -CA, and Fe -CA of pretreated hydrolysate improved (p < .05) by 17.41,
and Development Plan of Jiangsu Province, 20.00, 14.72, and 26.41%, respectively over nonsonicated hydrolysate. Amino acid con-
Grant/Award Numbers: BE2016355,
BE2016352 tent and hydrophobicity of SMP hydrolysate at the optimum sonication condition were
analyzed. Analyses indicated that ultrasonication could facilitate the releasing/unfolding of
hydrophobic amino acids from SMP over nonsonicated samples during enzymolysis with
high antioxidative capacity.
Practical applications
Sonication pretreatment has been presented to have notable impacts on the antioxi-
dant's capacities of protein hydrolysates. Sunflower-meal protein is an abundant and
low-cost residue of oil industries, and is considered a potential bioactive peptides
source, such as antioxidants. In this research, sonication pretreatment has been illus-
trated to be an effectual technique for the production of hydrolysates with antioxi-
dant capacity from SMP. So, optimization of ultrasonication conditions is vital to
examine the antioxidant capacities in SMP hydrolysates, which could be applied in
large-scale food system and future pharmaceutical activity research.
1 | INTRODUCTION presence of some diseases including diabetes, cancer, aging, hyperten-
sion, and Alzheimer's (Hajieva & Behl, 2006). Some artificial antioxidants
Antioxidant is commonly utilized to avoid lipid oxidation in food products including butylated hydroxytoluene and hydroxyanisole are widely uti-
to prevent the creation of undesirable flavors and odors and toxic com- lized to delay deterioration and discoloration of foods. Nevertheless, due
ponents (Lin & Liang, 2002). In addition, oxidative stress is implicated in to health hazards, utilization of these artificial antioxidants is limited
J Food Process Eng. 2019;e13084. wileyonlinelibrary.com/journal/jfpe © 2019 Wiley Periodicals, Inc. 1 of 12
https://doi.org/10.1111/jfpe.13084
2 of 12 DABBOUR ET AL.
(Park, Jung, Nam, Shahidi, & Kim, 2001). Therefore, there is growing of sunflower oil. The meals were ground by portable mill and passed
attention in search for antioxidants from natural and safe resources that through a 60-mesh sieve, then maintained in zip lock bags at tempera-
can improve the human body's antioxidant defenses to prevent oxidative ture of 4
C. All chemicals utilized in this work were of analytical-
damage (García-Moreno et al., 2014). grade:1,1-diphenyl-2-picrylhydrazyl (DPPH), iron (II) sulfate (FeSO4),
Using cheap agricultural residues to develop new foods has impor- hydrogen peroxide (H2O2), methanol, salicylic acid (C7H6O3), iron
tant economical and nutritional benefits. For example, sunflower meals, a (II) chloride (FeCl2), ferrozine (C20H13N4Na2O6S2), pyridine (C5H5N),
coproduct of oil extraction, because of its protein content up to 40–50% pyrocatechol violet (C19H14O7S) were bought from Sinopharm Chemical
(González-Pérez et al., 2002), has been recognized as a vital and low-cost Reagent Co., Ltd. (China). Enzyme alcalase 2.4 LFG (activity
raw material of protein for livestock feed and food ingredients. For 150,000 U/mL) was acquired from Novozymes Biotech. Co., Ltd. (China).
human benefits, huge efforts have been done to evolve efficient proce-
dures for preparation of acceptable products from sunflower meal (SM).
2.2 | Sonication pretreatments and enzymolysis
With rich protein content, therefore, enzymolysis of SMP is an appropri-
of SMP
ate procedure to produce bioactive peptides, which can be applied in
pharmaceutical and nutraceutical fields. In this respect, numerous sun- Dual-frequency ultrasound (Meibo Biotech. Co., China) was applied as
flower protein hydrolysates have revealed several bioactivities including the source of sonication in this experiment (detailed in our earlier
antihypertensive (Megías et al., 2004, 2009a), antihypocholesterolemic study [Dabbour, He, Ma, & Musa, 2018]). Sonication power (220 W),
(Megías et al., 2009b), antimicrobial (Taha, Mohamed, Wagdy, & time and dual-frequency (20/40 kHz) in a combined operating mode
Mohamed, 2013), and antioxidant (Megías et al., 2007; Taha et al., 2013). (pulsed on and off-time of 5 and 2 s, respectively) were controlled from
However, the poor solubility of sunflower proteins cause barrier the equipment control panel and temperature was controlled using a
between enzymes and proteins, leading to reduction in the affinity digital-thermostatic bath. The ultrasonic probes were submerged two
between them, which impactfully decrease the bioavailability of SMP centimeters into solution. The suspensions were pretreated at various
and restrict the release of functional peptides. Consequently, the devel- temperature (30–50
C), solvent-solid ratio (14–26 mL/g), and sonica-
opment of a more effective method to overcome these drawbacks is a tion time (10–30 min).
matter of great attention to food researchers. Following sonication treatment, pH of sunflower-meal protein
Ultrasonication technology is usually considered to be simple, cheap, solutions was adjusted to 9.0 using 1 mol/L sodium hydroxide. Subse-
environmentally friendly, and safe method (Zou et al., 2016), that make its quent to 15 min preheating in digital-thermostatic bath at tempera-
utilization have vital advantages over other technologies. Ultrasonic has ture of 50
C, enzyme alcalase was added (0.343 g/L) to initiate the
been applied for different purposes in food system such as changing enzymatic reaction. The pH value of solutions was maintained at
enzyme activity (Wei & Ye, 2011), facilitating enzymolysis (Dabbour, He, 9.0 by adding 1 mol/L sodium hydroxide through the enzymolysis
Mintah, Tang, &Ma, 2018), and improved biological properties of proteins (90 min). The reaction was inactivated after 90 min of hydrolysis by
(Yu et al., 2012). The cavitation impacts produced in solution by sonica- putting the hydrolysates in boiling water for 10 min. Afterward, the
tion waves could lead to degradation of proteins and changes in protein suspensions were subjected to centrifugation (4,500 g, 10 min) to
conformation, expose further hydrophobic sites and groups, additionally obtain supernatants, and kept at −20
C for further investigations. The
enhancing protein solubility and increasing the binding between matrix sonication parameters with various temperature, solvent-solid ratio,
and enzyme. This enhances the hydrolysis efficacy. To the best our and time combinations are displayed in Table 1. Control (classical
knowledge, no research on ultrasonic pretreatment of sunflower-meal enzymolysis) was carried out using an impeller-agitator device under
protein to prepare antioxidant hydrolysates has been reported. Therefore, the same parameters but without sonication. Under optimum condi-
the antioxidant capacities of sunflower-meal protein hydrolysate were tion, supernatant was frozen, lyophilized by a freeze dryer and stored
assessed comprehensively by DPPHSA, HRSA, Cu2+-CA, and Fe2+-CA. at −20
C to investigate the amino acid composition and surface
In this context, the purpose of the current research was to exam- hydrophobicity of SMP hydrolysate.
ine and optimize ultrasonic pretreatments using response-surface
methodology (RSM) to observe the highest antioxidant capacities for
2.3 | Experimental design
SMP hydrolysates. This work provided a theoretical basis to yield
sunflower-meal protein hydrolysates with better antioxidant activities. To maximize the antioxidants activity in sunflower protein hydroly-
The optimized hydrolysate could be applied as natural and safe anti- sates pretreated by ultrasonic, RSM technique was utilized. The best
oxidants for pharmacological and nutraceutical fields. sonication treatment was accessed by Box–Behnken's design (BBD).
Three-level-three-variable of temperature (
C, X1), solvent-solid ratio
(mL/g, X ), and sonication time (min, X ) were employed (Table 1). The
2 | MATERIALS AND METHODS 2 3
experimental design comprised of 17 experiments; and DPPHSA (%,
Y1), HRSA (%, Y2), Cu
2+-CA (%, Y3), and Fe
2+-CA (%, Y4) were selected2.1 | Materials and chemicals
as responses. Average values of these responses observed from three
SM, with 29.31% protein, which was utilized in our work was collected replicates were analyzed by Design-Expert 8.0.6.1 and fitted to a
from Xinjiang Jinhai Oils Co., Xinjiang, (China) as a residue of extraction quadratic-polynomial equation as:
DABBOUR ET AL. 3 of 12
TABLE 1 BBD for experimental factors (temperature, solvent-solid ratio and time) and the responses data
Run Temperature (
C) Solvent-solid ratio (mL/g) Time (min) DPPHSA (%) HRSA (%) Cu2+-CA (%) Fe 2+-CA (%)
1 40 20 20 55.86 64.21 45.50 41.96
2 40 20 20 58.33 65.50 47.28 40.58
3 40 26 10 48.88 55.30 38.40 29.85
4 40 20 20 57.75 66.11 46.91 39.50
5 50 14 20 37.09 80.40 45.50 42.62
6 30 14 20 37.69 55.90 41.90 30.50
7 50 20 30 40.87 65.00 52.37 46.33
8 40 14 30 37.70 77.49 50.74 42.30
9 50 20 10 49.06 59.75 39.95 29.10
10 30 20 30 48.48 74.71 45.57 32.81
11 40 20 20 56.40 65.95 48.20 40.22
12 30 20 10 54.83 50.15 38.20 24.13
13 40 20 20 58.87 64.90 47.50 41.32
14 30 26 20 38.81 72.76 40.13 27.20
15 40 26 30 41.01 59.12 38.20 36.72
16 50 26 20 37.99 55.84 35.21 35.83
17 40 14 10 44.37 47.48 32.66 31.67
X X X
Y = β + β X + β X2 + β X X ð1Þ samples were mixed with 6 mM FeSO4 (1 mL) and 1 mL of 6 mM H O ,0 i i ii i ij i j 2 2
agitated and incubated for 12 min. Then the resultant solution was
where Y represents predicted response; βi, βii, and βij signifies linear, mixed with 9 mM salicylic acid (1 mL), vortexed and allowed to react for


squared (second-order) and combination coefficients for independent 0.5 hr at temperature of 37 C. The absorbance of mixtures was read at
variables (X1, X2, and X3), β0 denotes intercept, and Xj and Xi rep- wavelength of 510 nm and distilled water (1 mL) in place of hydrolysate
resenting experimental factors. sample was utilized as a blank. HRSA (%) was determined as:
HRSA ð%Þ
2.4 | Antioxidants activity  Sample absorbance−Sample without salicylic acid absorbance
= 1− ×100
Blank absorbance
2.4.1 | DPPHSA assay ð3Þ
DPPH radical-scavenging activity was performed as outlined by Zhang, Li,
Miao, and Jiang (2011) with slight alterations (García-Moreno et al., 2014).
Suitable dilution of SMP hydrolysate (0.2 mL) was mixed with 3.4 mL of 2.4.3 | Fe2+-CA
0.1 mM DPPH radical dissolved in methanol (95%) and with 50 mM
2+
(pH 7.3) Tris buffer (0.40 mL), agitated and incubated (0.5 hr, 25
C) in dark. Fe -CA of hydrolysates was assessed as described by Zhang et al.
The absorbance of reactants was monitored using spectrophotometer at (2009) with slight modifications (García-Moreno et al., 2014). One mil-
wavelength of 517 nm against distilled water instead of SMP hydrolysate liliter of an appropriate dilution of hydrolysate suspension was added
(blank). Sample-control was done for every hydrolysate sample using to deionized water (3.5 mL). Thereafter, 2 mM FeCl2 (0.10 mL) was
added and subsequent to 3 min the resultant mixtures were mixed
methanol in place of DPPH. The DPPHSA (%) was estimated as:
with 5 mM ferrozine (0.20 mL) and then shaken vigorously. The absor-
Scavenging ð Þ bance was detected at wavelength of 562 nm after mixtures incuba- % 
Absorbance of sample−Absorbanceof sample control tion (20 min, 25

C), and distilled water (1 mL) instead of hydrolysate
= 1− − ×100
Absorbance of blank were applied as blank. Sample-control was done for every hydrolysate
ð2Þ but without ferrozine. The Fe2+-CA (%) was measured as in
Equation (2).
2.4.2 | HRSA 2.4.4 | Cu2+-CA
The HRSA was investigated according to Wang, Wang, Dang, Zheng, The technique of Zhu, Jie, Tang, and Xiong (2008) with alterations
and Zhang (2013) with some alterations. Aliquots (1 mL) of diluted was utilized to investigate Cu2+-CA of SMP hydrolysate. Briefly,
4 of 12 DABBOUR ET AL.
2 mM CuSO4 (1 mL) was added to 0.1% pyrocatechol violet (20 μL) TABLE 2 ANOVA (p-value, R2 and Adj.R2) of the fitted quadratic
and 10% pyridine (1 mL). After adding diluted sample (1 mL), the blue models for the examined parameters versus each response
color disappearance was observed by analyzing the absorbance after p-value
5 min of reaction at wavelength of 632 nm with spectrophotometer.
Source DPPHSA HRSA Cu2+-CA Fe2+-CA
Deionized water (equivalent volume) was applied for blank instead of
hydrolysate. Sample-control was done for every hydrolysate but with- Model <.0001 <.0001 .0004 <.0001
out pyrocatechol violet. Cu2+-CA of SMP hydrolysate (%) was mea- X1 .0360 .0928 .1750 <.0001
sured as in Equation (2). X2 .1289 .0021 .0056 .0033
X3 .0014 <.0001 .0001 <.0001
X X .9581 <.0001 .0401 .2577
2.5 | Amino acid analysis 1 2
X1X3 .6628 .0002 .1797 .0194
Composition of amino acid of sonicated and nonsonicated hydroly- X2X3 .7752 <.0001 .0010 .2260
sates at the optimum condition was investigated as outlined by X 21 .0002 .0351 .1942 .0004
Marino et al. (2010) with slight alterations. Hydrolysate of SMP for
X 22 <.0001 .2486 .0004 .0121
control and ultrasonication pretreatment was hydrolyzed in 6 mol/L
X 23 .0796 .0002 .0579 .0021
HCl for 24 hr at temperature of 110 ± 2
C in Pyrex microcapillary
Lack-of-fit .0895 .0638 .0691 .1155
tubes. Then, volume of resultant mixture was set to 50 mL with
R-squared .975 .991 .960 .979
deionized water. One milliliter of diluted hydrolysates was filtered
Adj R-squared .942 .978 .908 .952
(syringe filter 0.22 μm), lyophilized at temperature of 60
C using vac-
uum concentrator. Quantification was done with an amino acid ana- Note. p-value <.05 and <.01 are statistically significant and highly
lyzer (S-433D, Sykam GmbH Co., Germany). significant, respectively.
experimental values (Chen et al., 2018). The R2 and adjusted R2 values
2.6 | Surface hydrophobicity (H0) (Table 2) were 0.975 and 0.942, 0.991 and 0.978, 0.960 and 0.908,
The protocol of Kato and Nakai (1980) was utilized to estimate the and 0.979 and 0.952 for DPPHSA, HRSA, Cu
2+-CA, and Fe2+-CA of
hydrophobicity of hydrolysate with ANS as a probe of fluorescence. SMP hydrolysates, respectively, which demonstrated a strong correla-
SMP hydrolysate was dissolved in phosphate buffer (pH 8.0, tion degree among the predicted and actual values. In conclusion,
10 mmol/L) to attain different concentrations (0.04–0.1 mg/mL). Ali- ANOVA showed that the quadratic models are dependable for the pre-
quot (4 mL) of hydrolysate suspension was added to ANS (20 μL) diction of antioxidant capacities.
(8.0 mmol/L in the same buffer) and the resultant suspension was
kept (3 min) in dark. Fluorescence capacity (FC) of hydrolysate sus- 3.2 | Influence of experimental factors on the
pension was estimated at wavelength of 270 nm (excitation, slit antioxidant's activities of SMP hydrolysates
5.0 nm) and the scanning speed of 2 nm/s with a Cary Eclipse spectro-
photometer (Varian, Inc., Palo Alto). The emission spectrum was quan- 3.2.1 | DPPHSA of SMP hydrolysate
tified at 485 nm wavelength. The initial slope (H0) was determined It is reported that antioxidants function through various mechanisms.
using linear regression analysis.
So, it is suggested to use various procedures to assess antioxidant
capacity (Frankel & Meyer, 2000). In this work, DPPHSA, HRSA,
3 | RESULTS AND DISCUSSION Cu2+-CA, and Fe2+-CA of SMP hydrolysate treated by sonication
were evaluated and optimized (Table 1). To evaluate free-radical
3.1 | Model fitting scavengers of antioxidants, a broadly recognized procedure is the
DPPHSA (Li, Jiang, Zhang, Mu, & Liu, 2008). DPPH is a stable free-
In this investigation, the empirical models were established to examine
radical, which decolorizes when decreased by proton donor sub-
the impact of the interaction between various parameters (tempera-
stances, resulting in a decrease in absorptivity at wavelength of
ture, solvent-solid ratio, and sonication time) on the antioxidant's activ-
ities (DPPHSA, HRSA, Cu2+-CA, and Fe2+-CA) of SMP hydrolysate. To 517 nm. The reduction in color is an index of analyte scavenging
assess the adequacy of DPPHSA, HRSA, Cu2+-CA, and Fe2+-CA func- activity. The actual data of DPPHSA via different experimental fac-
tion in the experimental setup, the ultrasonication parameters were tors were displayed in Table 1. The DPPHSA ranged from 37.09 to
optimized. Additionally, the linear, second-order and interaction 58.87% and the maximum value of DPPHSA (58.87%) was found at
impacts of experimental factors were analyzed. Nevertheless, the suit- sonication temperature of 40

C, solvent-solid ratio of 20 mL/g and
ability of these quadratic models is confirmed with the determination- time of 20 min (run No. 13). The observed data were analyzed
coefficient (R2), adjusted R-squared, model significance and signifi- (regression analysis) to examine the impact of experimental factors
cance of lack-of-fit. The closer the adjusted R-squared and R-squared and their interactions on the DPPHSA (%). The observed coeffi-
values to unity, the more appropriate the empirical models fit the cients were utilized to create the following regression equation.
DABBOUR ET AL. 5 of 12
F IGURE 1 3D-response curves representing the combination impact of experimental parameters on DPPHSA (%) of SMP hydrolysates
Y1ðDPPHSAÞ= −198:44+5 
:62X1 + 14:16X2 + 0:73X3−0:0009X1X2 40 C, 20 mL/g and 20 min. DPPHSA (%) improved gradually with
−0:005X1X3−0:005X X −0:071X
2
2 3 −0:345X
2−0:02X2 increasing temperature. Yet, when the temperature was elevated to1 2 3
ð4Þ around 40
C, the DPPHSA reduced, as exhibited in Figure 1b. Figure 1c
exhibits combination effect of X2X3 and explained that DPPHSA was
Statistical analysis (analyzed by BBD) in Table 2 showed that the raised as solvent-solid ratio rises, whereas, when solvent-solid ratio was
DPPHSA model was remarkably significant (p < .01), while the lack-of-fit raised to around 20 mL/g, the DPPHSA rapidly reduced. Comparable
was nonsignificant (p = .0895), suggesting that this mathematical model result on the impact of temperature on DPPHSA has been indicated in
was fitwith strong prediction (R-squared =0.975; Adj.R-squared =0.942). literatures but for bioactive compound (Sood & Gupta, 2015; Tabaraki,
The linear impact of temperature and ultrasonication time (X1 and X3) Heidarizadi, & Benvidi, 2012).
were significant (p < .05)—positive effect on DPPHSA (%). Additionally,
the terms of X21 and X
2
2 had highly significant (p< .01)—negative corre- 3.2.2 | HRSA of SMP hydrolysate
lation on DPPHSA of SMP hydrolysate (Table 2 and Equation (4)),
whereas the other-terms were (p> .05) nonsignificant. Hydroxyl radical is highly reactive oxygen-radical that can react with vari-
Based on Equation (4), 3D-curves were plotted (Figure 1a–c) to char- ous biomolecules including DNA, carbohydrates, lipids, amino acids, pro-
acterize the influences of the experimental variables and their interaction tein, and nucleotide (You, Zhao, Regenstein, & Ren, 2010) and can lead to
on DPPHSA and to realize the best experimental point of the response. intense harm to cell tissues. Consequently, elimination of hydroxyl radical
Figure 1a illustrates the influence of the interaction of temperature and may be efficient defense versus different diseases in vivo. Evaluation of
solvent-solid ratio and clarified that maximal DPPHSA was observed at HRSA gives valuable information on antioxidants capacities. The observed
6 of 12 DABBOUR ET AL.
F IGURE 2 3D-response curves representing the combination impact of experimental parameters on HRSA (%) of SMP hydrolysates
data of HRSA (%) at various experimental variables were exhibited in Figure 2a–c shows the interaction impacts of temperature,
Table 1. The maximumHRSA value was 80.40% and found via experimen- solvent-solid ratio and sonication time on HRSA (%) of SMP hydroly-
tal parameters of 50
C, 14 mL/g and 20 min (run No. 5), while the mini- sates. Figure 2a displays that HRSA was obtained high at high levels
mumHRSA (47.48%) was found in combination of sonication temperature of solvent-solid ratio (X2) and temperature (X1) and the maximal value
of 30
C, solvent-solid ratio of 20 mL/g and time of 10 min (run No. 17). was obtained at low level of solvent-solid ratio (X2) and high level of
The fitted regression model of HRSA (%) is as follows: temperature (X1). Combined influence of temperature and sonication
time was exhibited in Figure 2b and clarified HRSA enhanced steadily
Y2ðHRSAÞ= −167:34+3:13X1 + 9:63X2 + 6:77X3−0:173X1X2 ð Þ by increasing sonication time. Moreover, Figure 2c illustrates mutual5
−0:048X1X3−0:109X2X3 + 0:017X
2 2 2
1 −0:023X2 −0:047X3 effect of X2X3 and explained that a higher sonication time (30 min)
and a lower solvent-solid ratio (14 mL/g) would provide a higher
The HRSA model showed a remarkably significant (p < .01), while HRSA of SMP hydrolysates. Similar finding on the influence of sonica-
the lack-of-fit illustrated a nonsignificant (p = .0638), confirming that tion time on HRSA has been reported in literature but for different
the current model was suitable to describe the actual HRSA data sample (Zou & Hou, 2017).
(R2 = 0.991 and Adj.R2 = 0.978) (Table 2). As displayed in Table 2,
solvent-solid ratio and sonication time exhibited highly significant
3.2.3 | Cu2+-CA of SMP hydrolysate
(p < .01)—positive correlation on HRSA. The impacts of X2 21 and X3
were significant (p< .05). Furthermore, all combined factors (X1X2, Transition metal-ions, including Cu
2+ and Fe2+, can accelerate the reac-
X1X3, and X2X3) presented a noticeably significant (p< .01)—negative tive oxygen species formation, leading to DNA harm and lipid oxidation
impact on HRSA of sunflower-meal protein hydrolysates. (Stohs & Bagachi, 1995). Particularly, Fe2+ produces hydroxyl-radical
DABBOUR ET AL. 7 of 12
F IGURE 3 3D-response curves representing the combination impact of experimental parameters on Cu2+-CA (%) of SMP hydrolysates
through Fenton reaction, thus accelerating the chain reaction of lipid oxi- prediction of Cu2+-CA values. Solvent-solid ratio (X2) and sonication
dation. Additionally, Fe2+ catalyzes the lipid peroxides decomposition, time (X3) were highly significant (p < .01) and had positive impact on
resulting in off-flavor creation. So, chelating of Cu2+ and Fe2+ with anti- Cu2+-CA of SMP hydrolysate. The effect of X22 was highly significant
oxidant could delay the oxidation reaction. The Cu2+-CA data (%) of SMP (p = .0004) and had negative influence on Cu2+-CA. Additionally,
hydrolysate of all 17 experimental runs were in the range of combination of temperature and solvent-solid ratio (X1X2) and
32.66–52.37% (Table 1). Run No. 17 (40
C, 14 mL/g and 10 min), solvent-solid ratio and sonication time (X2X3) had a significant effect
resulted in the lowest Cu2+-CA (%) value, while run No. 7 (50
C, 20 mL/g (p< .05)—negative correlation on Cu2+-CA.
and 30 min) provided the maximal value. Cu2+-CA results were analyzed 3D-curves plots were utilized to investigate the influence of
(multiple regressions) for fitting the regression model as follows: combination of temperature (X1), solvent-solid ratio (X2), and
  ultrasonication time (X ) on Cu2+3 -CA (%) of SMP hydrolysates
Y Cu2+3 −CA = −91:17+1:50X1 + 8:34X2 + 2:24X3−0:036X1X2 (Figure 3a–c). The interaction impacts of X1X2, X1X3, and X2X3 were
+0:013X X −0:076X X −0:012X2−0:145X2−0:019X2 examined. Figure 3a presents the combination impact of X1X2 and elu-1 3 2 3 1 2 3
ð6Þ cidated that Cu2+-CA was enhanced as temperature and solvent-solid
ratio rises, while, when solvent-solid ratio was raised to around
Statistical significance of the experimental factors and their com- 20 mL/g, the Cu2+-CA reduced. Moreover, temperature had a little
bination were displayed in Table 2. The Cu2+-CA model was highly influence on Cu2+-CA compared with ultrasonication time as exhibited
significant (p < .01) and correlated with the values of R2 (0.960) and in Figure 3b and the highest value of Cu2+-CA was observed at high
Adj. R2 (0.908), while lack-of-fit was nonsignificant (p = .0691). These levels of temperature (50
C) and sonication time (30 min). Combined
observations proved the mathematical model is satisfactory for influences of solvent-solid ratio and time is presented in Figure 3c
8 of 12 DABBOUR ET AL.
F IGURE 4 3D-response curves representing the combination impact of experimental parameters on Fe2+-CA (%) of SMP hydrolysates
which indicates a longer sonication time (30 min) and a lower solvent- (0.979) and Adj.R2 (0.952) values, but the lack-of-fit was insignificant
solid ratio (14 mL/g) would yield a greater Cu2+-CA. (p = .1155), demonstrating that this mathematical model could prop-
erly clarify the relation among the Fe2+-CA and the experimental vari-
2+ ables. Data analysis of Fe
2+-CA (Table 2) showed differences in the
3.2.4 | Fe -CA of SMP hydrolysates
temperature (X1), solvent-solid ratio (X2), and sonication time (X3) were
The impacts of temperature (X1), solvent-solid ratio (X2), and sonication highly significant (p < .01), showed positive influence on Fe2+-CA. Fur-
time (X3) on Fe
2+-CA (%) of SMP hydrolysate were examined by a BBD. thermore, three-quadratic terms (X21, X
2
2, and X
2
3) were significant
Results of Fe2+-CA (%) of all experimental runs increased from 24.13 to (p< .05)—negative impact on Fe2+-CA. Combination impacts of X1X3
46.33% (Table 1). Maximum value was found at the interaction of 50
C, was significant (p< .05), while another combination (X1X2 and X2X3)
20 mL/g and 30 min (run No. 7), but the lowest value was observed at illustrated nonsignificant (p> .05). For each term in the previous math-
30
C, 20 mL/g and 10 min (run No. 12). Relationship among Fe2+-CA (%) ematical models, a low p-value demonstrate a more significant influ-
and experimental variables (actual parameters) was determined as follows: ence on the experimental parameter (Quanhong & Caili, 2005).
  Figure 4a-c shows the interaction influences of X1X2, X1X3, and
Y Fe2+4 −CA = −91:85+3:84X1 + 3:12X2 + 1:31X3−0:015X1X2 X2X3 on Fe
2+-CA (%) of SMP hydrolysates. Figure 4a exhibits the com-
2 2 2++ 0:021X1X3−0:016X2X3−0:044X1 −0:064X2 −0:033X
2 bination impact among X1 and X2 and elucidated that Fe -CA was3
ð7Þ raised steadily as temperature (X1) rises, and the lesser temperature
(30
C) and a higher solvent-solid ratio (26 mL/g) provide a greater
The statistical significance of the Fe2+-CA model and the experi- Fe2+-CA. Moreover, maximal Fe2+-CA value was observed at high
mental factors were measured with p-value (Table 2). The model was levels of temperature and sonication time, while the least value was
highly significant (p < .01) for fitting the Fe2+-CA (%) with high R2 found at low levels of the same experimental variables (X1 and X3) as
DABBOUR ET AL. 9 of 12
TABLE 3 DPPHSA, HRSA, Cu2+-CA, and Fe2+-CA of SMP hydrolysate observed by optimal experimental variables of sonication and control
treatment
Responses (%)
Treatment DPPHSA HRSA Cu2+-CA Fe2+-CA
Ultrasonic pretreatment 49.92 ± 2.56 a 71.83 ± 3.12 a 51.02 ± 1.92 a 44.87 ± 3.02 a
Control 41.23 ± 1.85 b 57.46 ± 1.27 b 43.51 ± 2.47 b 33.02 ± 0.98 b
Increased rate (%) 17.41 20.00 14.72 26.41
Note. Mean data with various letters in column are significantly (p < .05).
displayed in Figure 4b. Mutual influences of solvent-solid ratio and 2016) and hydrophobic amino acids content. This was confirmed by
ultrasonication time is displayed in Figure 4c which demonstrates that the composition of amino acid and observation of hydrophobicity of
sonication time had significant impacts more than solvent-solid ratio SMP hydrolysate. Thus, an improvement in the alcalase activity to
on Fe2+-CA. facilitate the enzymolysis and then release more active-peptides with
high antioxidant capacities in suspension. Comparable observation
has been indicated in previous works (Abdualrahman et al., 2016;
3.3 | Validation of predictive optimum experimental
conditions Liang et al., 2017). Finally, sonication treatments over control were
more helpful for enzymolysis of sunflower-meal protein to prepare
The antioxidants activities of SMP hydrolysates were investigated at hydrolysates with high antioxidants activities.
the predicted optimum experimental variables to validate the devel-
oped models observed from RSM. This technique was applied in our
3.5 | Amino acid analysis
study to access the best experimental conditions which could achieve
the maximal antioxidants activities. The predicted levels of tempera- Composition of amino acid was quantified for sonicated and non-
ture 38.68
C, solvent-solid ratio 20.38 mL/g and time 11.49 min gave sonicated hydrolysates (Table 4) to examine the influences of
the highest DPPHSA (59.65%), and the maximal HRSA (82.26%) was ultrasonication on the antioxidant potential of SMP hydrolysates.
realized under the predicted experimental parameters of 48.64
C, From Table 4, the untreated and pretreated samples were rich in Glu
14.12 mL/g and 29.73 min. Greatest value of Cu2+-CA (53.75%) was and Asp. Contents of hydrophobic amino acid (HAA) raised by 8.23%,
achieved at 47.81
C, 14.21 mL/g, and 29.25 min, while the experi-
mental factors of 47.77
C, 14.51 mL/g and 29.47 min provided maxi- TABLE 4 Composition of amino acid (%) of nonsonicated and
2+ sonicated SMP hydrolysatemum Fe -CA value (45.25%). The optimal experimental point for all
experimental responses was at 42.50
C, 18.16 mL/g and 26.52 min. Amino acids Control Ultrasonic pretreatment
Under this condition, the predicted DPPHSA, HRSA, Cu2+-CA and Fe2 Asp 12.6 ± 0.19 12.0 ± 0.29
+-CA values were 52.09, 70.05, 50.85, and 43.35% respectively. Thr 2.9 ± 0.06 3.5 ± 0.15
These values are verified with the obtained results. The observed Ser 3.7 ± 0.04 3.4 ± 0.09
DPPHSA of 49.92%, HRSA of 71.83%, Cu2+-CA of 51.02%, and Fe2 Glu 20.1 ± 0.62 22.2 ± 0.49
+-CA of 44.87% were insignificant (p > .05) with the predicted results. Pro 3.7 ± 0.09 3.6 ± 0.13
This powerfully demonstrates that the regression model is appropriate Gly 6.0 ± 0.21 5.4 ± 0.11
to predict DPPHSA, HRSA, Cu2+-CA, and Fe2+-CA at selected experi- Ala 4.3 ± 0.17 4.9 ± 0.08
mental variables.
Cys 1.6 ± 0.02 1.6 ± 0.04
Val 3.6 ± 0.18 3.4 ± 0.15
3.4 | Comparison of sonication and control Met 2.4 ± 0.03 2.7 ± 0.05
pretreatments Ile 2.2 ± 0.07 3.0 ± 0.10
Leu 6.5 ± 0.25 6.7 ± 0.31
The results of antioxidant activity of SMP hydrolysates pretreated
with sonication and control were compared (Table 3). Sonication at Tyr 2.5 ± 0.10 2.2 ± 0.04
optimal treatment point remarkably raised (p < .05) antioxidant capac- Phe 4.4 ± 0.18 4.9 ± 0.11
ities over control. The values of DPPHSA, HRSA, Cu2+-CA, and Fe2 His 2.7 ± 0.09 2.9 ± 0.08
+-CA of sonicated SMP hydrolysates enhanced (p < .05) by 17.41, Lys 1.4 ± 0.02 1.6 ± 0.05
20.00, 14.72, and 26.41%, respectively compared with control. These Arg 3.5 ± 0.13 3.5 ± 0.07
phenomena may be attributed to sonication pretreatment motivated HAA (%) 30.0 32.7
alterations to protein structure, resulting in a loosened molecular Note. HAA, hydrophobic amino acids (Val, Thr, Met, Ala, Ile, Phe, Pro,
structure and rise in proteins surface-hydrophobicity (Zhang et al., and Leu).
10 of 12 DABBOUR ET AL.
4 | CONCLUSIONS
In this observation, RSM was efficaciously utilized for accessing the
experimental variables for optimal antioxidants activities from SMP
hydrolysates. Temperature, solvent-solid ratio and sonication time condi-
tions significantly and positively affected DPPHSA, HRSA, Cu2+-CA, and
Fe2+-CA. ANOVA presented a high R2 and Adj.R2 0.975 and 0.942 for
DPPHSA, 0.991 and 0. 978 for HRSA, 0.960 and 0.908 for Cu2+-CA and
0.979 and 0.952 for Fe2+-CA, confirming a reasonable fit of the regres-
sion models with the measured data. Optimal experimental conditions
for antioxidants activities were at 42.50
C, 18.16 mL/g and 26.52 min.
This point provided DPPHSA of 52.09%, HRSA of 70.05%, Cu2+-CA of
50.85% and Fe2+-CA of 43.35%. For these responses, experimental data
were noticed to be close to the predicted data realized by the quadratic
models. At optimum sonication condition, antioxidant capacities of soni-
cated hydrolysate were improved (p < .05) over nonsonicated one. This
F IGURE 5 Surface hydrophobicity of nonsonicated (C) and
enhancement concurred with the increase in the hydrophobic amino
sonicated (S) SMP hydrolysate
acids content and hydrophobicity of sonicated hydrolysate. According to
the findings observed in this work, these results could be scaled to indus-
especially remarkable increase in Ile (26.67%) with sonication over tries to produce antioxidants from SMP hydrolysates in large quantities.
nonsonicated hydrolysate. The finding is in accordance with the
observation of Huang, Dai, Li, and Ma (2015). This phenomenon of ACKNOWLEDGMENTS
the sonicated samples inclines to form peptides with C-terminal HAA
We are grateful for the financial support provided by Primary
(Ala, Val, Leu, Ile, Phe, Pro, Thr, and Met) more than untreated one
Research and Development Plan (BE2016352, BE2016355), National
during hydrolysis, since sonication results in more hydrophobic
Primary Research and Development Plan of Jiangsu Province
regions/sites and groups inside the molecules to be exposed to sur-
(2016YFD0401401) and sponsored by Qing Lan Project.
rounding surface. Hydrolysates with high antioxidant activity are asso-
ciated with HAA. Thus, more hydrophobicity may cause more
antioxidant capacities of sonicated hydrolysates. Our findings NOTATION
suggested ultrasonication could be easy to produce hydrolysate with BBD Box–Behnken's design
strong antioxidant potential. Cu2+-CA Cu2+ chelating activity (%)
DPPHSA DPPH-scavenging activity (%)
Fe2+-CA Fe2+ chelating activity (%)
3.6 | Surface hydrophobicity HAA hydrophobic amino acids
To evaluate the conformational alterations in protein, hydrophobicity H0 surface hydrophobicity
is important. It shows the groups on periphery of protein molecules HRSA hydroxyl-radical scavenging activity (%)
(Chandrapala, Zisu, Palmer, Kentish, & Ashokkumar, 2011), that influ- RSM response-surface methodology
SM sunflower meal
ences on protein functionality. Impact of ultrasonication pretreatment
SMP sunflower-meal protein
at the optimum points on H0 of SMP hydrolysates was presented in
X temperature (
C)
Figure 5. Compared with control, H0 of the sonicated SMP hydrolysate
1
X2 solvent-solid ratio (mL/g)
was improved (144.68 ± 12.83–239.55 ± 6.55, p < .05). Our observa-
X3 sonication time (min)
tion agreed with Xiong et al. (2018). The improvement in H0 of hydro-
Y1 DPPHSA (%)
lysate indicates that SMP hydrolysate molecules were stretched
Y2 HRSA (%)
subsequent to ultrasonication. This may be attributed to the
Y Cu2+3 -CA (%)
microstreaming and cavitation actions produced by sonication, leading
Y4 Fe
2+-CA (%)
to a rise in number of hydrophobic groups and sites/regions buried at
the interior of hydrolysate; thus exposing the polar sites (Jakopovic,
Barukcic, & Božanic, 2016), and then enhancing the hydrophobicity of
GREEK SYMBOLS
SMP hydrolysate. Combined with the findings of amino acids composi-
tion, these observations may offer a clarification for the obtained βi linear coefficients for independent variables
enhancement in antioxidant capacities of sonicated SMP hydrolysates. βii second-order coefficients for independent variables
DABBOUR ET AL. 11 of 12
and antioxidant properties of corn protein hydrolysates. Journal of
βij combination coefficients for independent variables Food Quality, 2017, 1–10.
β0 Intercept Lin, C.-C., & Liang, J.-H. (2002). Effect of antioxidants on the oxidative sta-
bility of chicken breast meat in a dispersion system. Journal of Food
ORCID Science, 67, 530–533.
Marino, R., Iammarino, M., Santillo, A., Muscarella, M., Caroprese, M., &
Mokhtar Dabbour https://orcid.org/0000-0001-6109-9065 Albenzio, M. (2010). Technical note: Rapid method for determination
Ronghai He https://orcid.org/0000-0002-0904-0522 of amino acids in milk. Journal of Dairy Science, 93(6), 2367–2370.
Megías, C., Del Mar Yust, M., Pedroche, J., Lquari, H., Girón-Calle, J.,
Alaiz, M., … Vioque, J. (2004). Purification of an ace inhibitory peptide
after hydrolysis of sunflower (Helianthus annuus L.) protein isolates.
REFERENCES
Journal of Agricultural and Food Chemistry, 52(7), 1928–1932.
Abdualrahman, M. A. Y., Ma, H., Zhou, C., Yagoub, E. A., Hu, J., & Yang, X. Megías, C., Pedroche, J., Yust, M. d. M., Alaiz, M., Girón-Calle, J.,
(2016). Thermal and single frequency counter-current ultrasound pre- Millán, F., & Vioque, J. (2009a). Purification of angiotensin converting
treatments of sodium caseinate: Enzymolysis kinetics and thermody- enzyme inhibitory peptides from sunflower protein hydrolysates by
namics, amino acids composition, molecular weight distribution and reverse-phase chromatography following affinity purification. LWT—
antioxidant peptides. Journal of the Science of Food and Agriculture, 96 Food Science and Technology, 42(1), 228–232.
(15), 4861–4873. Megías, C., Pedroche, J., Yust, M. d. M., Alaiz, M., Girón-Calle, J.,
Chandrapala, J., Zisu, B., Palmer, M., Kentish, S., & Ashokkumar, M. (2011). Millán, F., & Vioque, J. (2009b). Sunflower protein hydrolysates reduce
Effects of ultrasound on the thermal and structural characteristics of cholesterol micellar solubility. Plant Foods for Human Nutrition, 64(2),
proteins in reconstituted whey protein concentrate. Ultrasonics 86–93.
Sonochemistry, 18, 951–957. Megías, C., Pedroche, J., Yust, M. M., Girón-Calle, J., Alaiz, M., Millán, F., &
Chen, S., Zeng, Z., Hu, N., Bai, B., Wang, H., Suo, Y., & Chen, S. (2018). Vioque, J. (2007). Affinity purification of copper-chelating peptides
Simultaneous optimization of the ultrasound-assisted extraction for from sunflower protein hydrolysates. Journal of Agricultural and Food
phenolic compounds content and antioxidant activity of Lycium Chemistry, 55(16), 6509–6514.
ruthenicum Murr. Fruit using response surface methodology. Food Park, P. J., Jung, W. K., Nam, K. S., Shahidi, F., & Kim, S. K. (2001). Purifica-
Chemistry, 242, 1–8. tion and characterization of antioxidative peptides from protein hydro-
Dabbour, M., He, R., Ma, H., & Musa, A. (2018). Optimization of ultrasound lysate of lecithin-free egg yolk. Journal of the American Oil Chemists'
assisted extraction of protein from sunflower meal and its physico- Society, 78(6), 651–656.
chemical and functional properties. Journal of Food Process Engineering, Quanhong, L., & Caili, F. (2005). Application of response surface methodol-
41(5), 1–11. ogy for extraction optimization of germinant pumpkin seeds protein.
Dabbour, M., He, R., Mintah, B., Tang, Y., & Ma, H. (2018). Ultrasound Food Chemistry, 92, 701–706.
assisted enzymolysis of sunflower meal protein: Kinetics and thermo-
Sood, A., & Gupta, M. (2015). Extraction process optimization for bioactive
dynamics modeling. Journal of Food Process Engineering, 41(7), 1–10.
compounds in pomegranate peel. Food Bioscience, 12, 100–106.
Frankel, E. N., & Meyer, A. S. (2000). The problems of using one-
Stohs, S. J., & Bagachi, D. (1995). Oxidative mechanisms in the toxicity of
dimensional methods to evaluate multifunctional food and biological
metal ions. Free Radical Biology & Medicine, 18(2), 321–336.
antioxidants. Journal of the Science of Food and Agriculture, 80(13),
1925–1941. Tabaraki, R., Heidarizadi, E., & Benvidi, A. (2012). Optimization of
ultrasonic-assisted extraction of pomegranate (Punica granatum L.)
García-Moreno, P. J., Batista, I., Pires, C., Bandarra, N. M., Espejo-
peel antioxidants by response surface methodology. Separation and
Carpio, F. J., Guadix, A., & Guadix, E. M. (2014). Antioxidant activity of
Purification Technology, 98, 16–23.
protein hydrolysates obtained from discarded Mediterranean fish spe-
cies. Food Research International, 65, 469–476. Taha, F. S., Mohamed, S. S., Wagdy, S. M., & Mohamed, G. F. (2013). Anti-
González-Pérez, S., Merck, K. B., Vereijken, J. M., Van Koningsveld, G. A., oxidant and antimicrobial activities of enzymatic hydrolysis products
Gruppen, H., & Voragen, A. G. J. (2002). Isolation and characterization from sunflower protein isolate. World Applied Sciences Journal, 21(5),
of undenatured chlorogenic acid free sunflower (Helianthus annuus) 651–658.
proteins. Journal of Agricultural and Food Chemistry, 50(6), 1713–1719. Wang, J., Wang, Y., Dang, X., Zheng, X., & Zhang, W. (2013). Housefly lar-
Hajieva, P., & Behl, C. (2006). Antioxidants as a potential therapy against vae hydrolysate: Orthogonal optimization of hydrolysis, antioxidant
age-related neurodegenerative diseases: Amyloid Beta toxicity and activity, amino acid composition and functional properties. BMC
Alzheimer's disease. Current Pharmaceutical Design, 12, 699–704. Research Notes, 6(1), 197–206.
Huang, L., Dai, C., Li, Z., & Ma, H. (2015). Antioxidative activities and pep- Wei, Y., & Ye, X. (2011). Effect of 6-benzylaminopurine combined with
tide compositions of corn protein hydrolysates pretreated by different ultrasound as pre-treatment on quality and enzyme. Journal of Food
ultrasonic methods. Journal of Food and Nutrition Research, 3(7), Processing and Preservation, 35, 587–595.
415–421. Xiong, T., Xiong, W., Ge, M., Xia, J., Li, B., & Chen, Y. (2018). Effect of high
Jakopovic, K. L., Barukcic, I., & Božanic, R. (2016). Physiological signifi- intensity ultrasound on structure and foaming properties of pea pro-
cance, structure and isolation of α-lactalbumin. Mljekarstvo, 66 tein isolate. Food Research International, 109, 260–267.
(1), 3–11. You, L., Zhao, M., Regenstein, J. M., & Ren, J. (2010). Purification and identifi-
Kato, A., & Nakai, S. (1980). Hydrophobicity determined by a fluorescence cation of antioxidative peptides from loach (Misgurnus anguillicaudatus)
probe method and its correlation with surface properties of proteins. protein hydrolysate by consecutive chromatography and electrospray
Biochimica et Biophysica Acta (BBA)—Protein Structure, 624(1), 13–20. ionization-mass spectrometry. Food Research International, 43(4),
Li, Y., Jiang, B., Zhang, T., Mu, W., & Liu, J. (2008). Antioxidant and free 1167–1173.
radical-scavenging activities of chickpea protein hydrolysate (CPH). Yu, L., Sun, J., Liu, S., Bi, J., Zhang, C., & Yang, Q. (2012). Ultrasonic-
Food Chemistry, 106(2), 444–450. assisted enzymolysis to improve the antioxidant activities of peanut
Liang, Q., Ren, X., Ma, H., Li, S., Xu, K., & Oladejo, A. O. (2017). Effect of (Arachin conarachin L.) antioxidant hydrolysate. International Journal of
low-frequency ultrasonic-assisted enzymolysis on the physicochemical Molecular Sciences, 13(7), 9051–9068.
12 of 12 DABBOUR ET AL.
Zhang, J., Zhang, H., Wang, L., Guo, X., Wang, X., & Yao, H. (2009). Antioxidant protein with single-frequency countercurrent and pulsed ultrasound-
activities of the rice endosperm protein hydrolysate: Identification of the assisted processing. Ultrasonics Sonochemistry, 28, 294–301.
active peptide. European Food Research and Technology, 229(4), 709–719. Zou, Y., & Hou, X. (2017). Sonication enhances quality and antioxidant
Zhang, T., Li, Y., Miao, M., & Jiang, B. (2011). Purification and characterisa- activity of blueberry juice. Food Science and Technology (Campinas),
tion of a new antioxidant peptide from chickpea (Cicer arietium L.) pro- 37(4), 599–603.
tein hydrolysates. Food Chemistry, 128(1), 28–33.
Zhang, Y., Wang, B., Zhou, C., Atungulu, G. G., Xu, K., Ma, H., …
Abdualrahman, M. A. Y. (2016). Surface topography, nano-mechanics
and secondary structure of wheat gluten pretreated by alternate dual-
frequency ultrasound and the correlation to enzymolysis. Ultrasonics How to cite this article: Dabbour M, He R, Mintah B, Ma H.
Sonochemistry, 31, 267–275. Antioxidant activities of sunflower protein hydrolysates
Zhu, L., Jie, C., Tang, X., & Xiong, Y. L. (2008). Reducing, radical scavenging, treated with dual-frequency ultrasonic: Optimization study.
and chelation properties of in vitro digests of alcalase-treated zein hydro-
J Food Process Eng. 2019;e13084. https://doi.org/10.1111/
lysate. Journal of Agricultural and Food Chemistry, 56(8), 2714–2721.
Zou, Y., Ding, Y., Feng, W., Wang, W., Li, Q., Chen, Y., … Wu, X. (2016). jfpe.13084
Enzymolysis kinetics, thermodynamics and model of porcine cerebral