Food Chemistry 136 (2013) Contents lists available at SciVerse ScienceDirect. Food Chemistry

Transcription

1 Food Chemistry 136 (213) Contents lists available at SciVerse ScienceDirect Food Chemistry journal homepage: Screening of whey protein isolate hydrolysates for their dual functionality: Influence of heat pre-treatment and enzyme specificity Randy Adjonu a,b, Gregory Doran a,b, Peter Torley a,b,c,, Samson Agboola a,b a EH Graham Centre for Agricultural Innovation, Charles Sturt University, Private Bag 588, Wagga Wagga, NSW 2678, Australia b School of Agricultural & Wine Sciences, Charles Sturt University, Private Bag 588, Wagga Wagga, NSW 2678, Australia c National Wine & Grape Industry Centre, Charles Sturt University, Private Bag 588, Wagga Wagga, NSW 2678, Australia article info abstract Article history: Received 14 May 212 Received in revised form 4 September 212 Accepted 14 September 212 Available online 23 September 212 Keywords: Bioactive peptides Whey protein isolate hydrolysates Free radical scavenging activity (FRSA) Oxygen radical absorbance capacity (ORAC) Surface hydrophobicity Dual functional peptides Heat pre-treated and non heat pre-treated whey protein isolate (WPI) were hydrolysed using a-chymotrypsin (chymotrypsin), pepsin and trypsin. The in vitro antioxidant activity, ACE-inhibition activity and surface hydrophobicities of the hydrolysates were measured in order to determine if peptides with dual functionalities were present. Dual functional peptides have both biological (e.g. antioxidant, ACE-inhibition, opioid activities) and technological (e.g. nanoemulsification abilities) functions in food systems. Heat pre-treatment marginally enhanced the hydrolysis of WPI by pepsin and trypsin but had no effect on WPI hydrolysis with chymotrypsin. With the exception of the hydrolysis by trypsin, heat pretreatment did not affect the peptide profile of the hydrolysates as analysed using size exclusion chromatography, or the antioxidant activity (P >.5). Heat pre-treatment significantly affected the ACE-inhibition activities and the surface hydrophobicities of the hydrolysates (P <.5), which was a function of the specificity of the hydrolysing enzyme. Extended hydrolysis (up to 24 h) had no significant effect on the DH and the molecular weight profiles (P >.5) but in some instances caused a reduction in the antioxidant activity of WPI hydrolysates. The chymotrypsin hydrolysate showed a broad MW size range, and was followed by pepsin and then trypsin. The bioactivities of the hydrolysates generally decreased in the order; chymotrypsin > trypsin > pepsin. This study showed that by manipulating protein conformation with pre-hydrolysis heat treatment, combined with careful enzyme selection, peptides with dual functionalities can be produced from WPI for use as functional ingredients in the manufacture of functional foods. Ó 212 Elsevier Ltd. All rights reserved. 1. Introduction Peptides from milk protein sources are known to have antioxidant activity and ACE inhibitory activity that help protect the human body against negative effects of free radicals and to maintain a normal blood pressure, respectively (Korhonen & Pihlanto, 26). The potential of these peptides in human nutrition management is increasingly being acknowledged (Lee, Skurk, Hennig, & Hauner, 27). These peptides have simple structures, and are considered safe and healthy compounds which are easily absorbed by the human body (Li, Le, Shi, & Shrestha, 24). Enzymatic hydrolysis of food proteins produces peptides, including antioxidative and ACE-inhibitory peptides (Korhonen & Pihlanto, 26). Enzymatic hydrolysis has also proven to be the most promising method of producing bioactive peptides from Corresponding author at: School of Agricultural & Wine Sciences, Charles Sturt University, Private Bag 588, Wagga Wagga, NSW 2678, Australia. Tel.: address: ptorley@csu.edu.au (P. Torley). proteins, and in many instances, has been shown to enhance the bioactivity of intact proteins (Pihlanto, 26). Since enzymes are usually highly specific in their mode of action, careful enzyme selection means they can be used to produce hydrolysates suitable for different food applications. Peptide bioactivity is largely dependent on the specificity of the enzyme used, the protein source and any treatment prior to hydrolysis that modifies the native protein structure (e.g. heat treatment) (Gauthier & Pouliot, 23). Heat pre-treatment of proteins prior to enzymatic hydrolysis results in intra- and inter-molecular disulphide interchanges and possible protein conformational changes (Lee, Morr, & Ha, 1992), which will partly dictate the nature of peptides released during hydrolysis, and hence their resultant functionalities. Heat may also induce non-specific peptide bond cleavage and enhance the hydrolysis of some globular proteins, such as b-lactoglobulin (Cheison, Schmitt, Leeb, Letzel, & Kulozik, 21). The effect of heat pretreatment on the structural technological functionalities of milk proteins and their peptides have been extensively addressed over the past decades (Dissanayake & Vasiljevic, 29; Lee et al., 1992), whereas the influence of heat pre-treatment on the /$ - see front matter Ó 212 Elsevier Ltd. All rights reserved.

2 1436 R. Adjonu et al. / Food Chemistry 136 (213) biological functionalities of milk proteins have received limited attention. Heat pre-treatment prior to hydrolysis could therefore be used to alter the proportions of different multifunctional peptides generated during enzymatic hydrolysis of milk proteins. Whey proteins constitute about 2% of the total proteins in milk. With approximately 86 million tonnes produced per year as by products from cheese manufacture, they are readily available, have a high nutritional value owing to a relatively high content of essential amino acids (Custódio et al., 29), and are easily digested by gastrointestinal enzymes such as chymotrypsin, pepsin and trypsin. The structural and technological functionalities of whey proteins have been extensively studied and well characterised, and have consequently been used as a primary substrate for the generation of multifunctional food peptides in the food industry (Madureira, Tavares, Gomes, Pintado, & Malcata, 21). Whey proteins have also been shown to contain various bioactive peptides e.g. antihypertensive, antithrombotic, opiate peptides (Lourenco da Costa, Antonio da Rocha Gontijo, & Netto, 27; Madureira et al., 21; Pihlanto-Leppälä, Koskinen, Piilola, Tupasela, & Korhonen, 2). These peptides can be released during in vivo digestion with enzymes such as chymotrypsin, pepsin and trypsin as well as in vitro enzymatic hydrolysis, and once they are released, they may act as regulatory compounds with hormone-like effects (Korhonen & Pihlanto, 26). In vitro digestion offers high controllability over the hydrolysis process and hence can be optimised in order to generate multifunctional peptides for possible inclusion in the manufacture of functional foods. With the growing interest in functional bioactive ingredients from food protein hydrolysate sources, it is necessary to identify peptides that have technological functions in food systems (e.g. nanoemulsification abilities) and also have biological function when consumed (e.g. antioxidant, ACE-inhibition, opioid activities). These peptides may be capable of stabilising nanoemulsions due to the particularly small droplet sizes of this category of emulsions. Thus, in addition to their bioactivities, the emulsifying potentials of these peptides could be exploited in food nanoemulsion systems. The objective of this work was to generate peptides from WPI that would have their functionality assessed to determine their potential use as dual functional ingredients in food nanoemulsion systems. In this paper, the effect of hydrolysis conditions (heat pre-treatment, enzyme specificity, time) on peptide formation (shown by the degree of hydrolysis) is discussed. The peptides generated were assayed to determine their potential as dual functional ingredients with both technological (molecular weight by size exclusion chromatography [SEC] and surface hydrophobicity and as indicators of emulsifying potential) and biological (antioxidant and ACE-inhibitory properties) functionalities. In addition to obtaining peptides potentially similar to those from in vivo action in the gut, this study is aimed at generating both biologically and technologically dual functional peptides for inclusion as functional ingredients in functional food manufacture. 2. Materials and methods 2.1. Materials Whey protein isolate (WPI) was purchased from MyoPure Pty Ltd (Petersham, Australia). The digestion enzymes (chymotrypsin, pepsin and trypsin), 2,2 -azinobis-(3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt (ABTS), 6-hydroxy-2,5,7,8-tetramethylchroman-2-acid (Trolox), o-phthaldialdehyde (OPA), dithiothreitol (DTT), bovine serum albumin, ovalbumin, b-lactoglobulin A, cytochome C and cytidine were purchased from Sigma Aldrich (Sydney, Australia). Fluorescein (FL) and 2,2 -azobis(2-methylpropionamidine) dihydrochloride (AAPH) were obtained from Merck Pty Ltd (Melbourne, Australia). All other chemicals were of reagent grade Preparation of heat pre-treated WPI WPI at a concentration of 5% (w/v) was heated at 8 C for 15 min in deionised water. After heating, the suspension was allowed to cool down to room temperature and concentrated by rotary evaporation (Rotavapor R-21 fitted with a vacuum controller V-85, Buchi Labortechnik, Flawil, Switzerland) at 37 C. The concentrates were freeze dried, homogenised using a blender and then stored at 2 C for further analyses Enzymatic hydrolysis WPI hydrolysates were prepared by suspending either the heat pre-treated or non heat pre-treated WPI in 1 mm phosphate buffer, ph 7, at a concentration of 5% (w/v). The suspension was then stirred and allowed to hydrate and equilibrate to the working temperature (37 C) for 3 min. The suspensions were then adjusted to enzymes working ph (ph 2.6 for pepsin and 7.8 for chymotrypsin and trypsin) with either 2 M HCl or NaOH, respectively. Enzymes were then added at an enzyme: substrate ratio of 1:4 with ph monitoring. The system was stirred throughout the hydrolysis process to avoid sedimentation especially with the heat pre-treated samples. The ph was maintained at the working ph values with 2. M HCl (pepsin) or NaOH (chymotrypsin and trypsin) for the first 12 h and left overnight for the 24-h hydrolysis. After the hydrolysis, the ph was adjusted to neutrality with 2. M NaOH or 2. M HCl accordingly. The hydrolysates were then heated at 9 C for 15 min to inactivate the enzymes. The suspensions were allowed to cool down to ambient temperature and then freeze dried or kept at 2 C for further analyses. The degree of hydrolysis (DH) was determined by the OPA method as described by Nielsen, Petersen, and Dambmann (21) Size exclusion chromatography Molecular weight (MW) distribution of hydrolysates was analysed by gel filtration chromatography under isocratic conditions using a Shodex Protein KW-82.5 column (8. mm 3 mm) fitted with a Protein-Pak 125 Sentry Guard Column (Waters Pty, Sydney, Australia) on an HPLC system. Hydrolysates were dissolved in phosphate buffer saline (5 mm Na 2 HPO 4 /NaH 2 PO 4 and 15 mm Na 2 SO 4, ph 7.) to prepare a 1 mg/ml solution and filtered through.22 lm syringe filter with 1 ll being injected into the column. Elution was at room temperature,.8 ml/min flow rate and peak absorbance was monitored at 214 nm. Each sample was run in triplicate. Bovine serum albumin (66. kda), ovalbumin (44.3 kda), b-lactoglobulin A (18.3 kda), cytochome C (12.4 kda) and cytidine (.243 kda) were run as standards. The percentage abundance (area under peak) of the determined molecular weight was obtained from the HPLC software (Varian STAR chromatography workstation version 6.41, Varian Inc., Victoria, Australia) Biological activities Antioxidant activity Free radical scavenging activity (FRSA). The free radical scavenging activity was carried out using the ABTS decolourising assay according to the method by Re et al. (1999). Briefly, ABTS (7 mm) and potassium persulphate (2.45 mm) were reacted together in 1 mm phosphate buffer saline (PBS, ph 7.4) to generate ABTS +. The reaction solution was kept at the dark for up to 14 h before use. Before each determination, the free radical mixture was diluted with the working buffer to an absorbance of.7 ±.2 at

3 R. Adjonu et al. / Food Chemistry 136 (213) nm. The reaction mixture (1.7 ml) contained 1.5 ml ABTS + and 2. ml of hydrolysates solution (1 5 mg/l) or trolox (5 25 lm). The mixture was then shaken vigorously for 3 s, incubated at 3 C for 6 min, and the final absorbance was read at 734 nm. Each sample was run in triplicate and PBS was run as blank. The percentage inhibition of the blank corrected absorbance was computed for each sample and trolox at the varying concentrations according to the formula below: % Inhibition ¼ Absorbance ðt¼þ Absorbance ðt¼6þ Absorbance ðt¼þ where absorbance t = is absorbance of ABTS + before addition of antioxidant (.7 ±.2) and absorbance t = 6 is absorbance after 6 min incubation period. The final antioxidant activity was expressed in trolox equivalent (lmol TE/mg protein sample) by dividing the slope of the linear regression fit of % inhibition versus concentration of the hydrolysates by that of the trolox standard regression fit according to the formula: Antioxidant activity ¼ Slope for each sampleðmg 1 LÞ Slope for trolox standardðlmol 1 LÞ Oxygen radical absorbance capacity (ORAC). ORAC assay was performed according to the method by Hernández-Ledesma, Davalos, Bartolome, and Amigo (25), with slight modifications. Briefly, the final assay mixture (2 ll) contained 15 ll of 7 nm fluorescein (FL), 25 ll antioxidant [trolox ( lm) as a standard or sample ( mg/l)] and 25 ll of 36 mm AAPH in 75 mm phosphate buffer (ph 7.4). Fluorescence data (excitation = 485 nm, emission = 52 nm) were collected using a FLUOstar Omega plate reader (BMG, GmbH, Offenburg, Germany) over 12 min at physiological temperature. The plate reader was equipped with two reagent injectors, an incubator (6 C) and top and bottom optics readers. Injector 1 was programmed to dispense 15 ll of FL during the second cycle and injector 2 dispensed 25 ll of AAPH during the fourth cycle. Prior to the first injection, the microplate with the antioxidants were incubated at the working temperature (37 C) for 2 min. All reaction mixtures were prepared in duplicate and at least three independent runs were performed for each sample. Fluorescence measurements obtained using the bottom optics, were exported to Microsoft Excel for further analysis. The area under the fluorescence decay curve (AUC) was calculated according to: i¼12 AUC ¼ 1 þ X i¼4 f i f 4 where f i = fluorescence reading at time i and f 4 = initial reading after addition of AAPH. The net AUC was calculated according to: NetAUC ¼ AUC ðantioxidantþ AUC ðblankþ The linear regression fit between the net AUC and the antioxidant concentration was obtained for each sample. The ORAC value (lmol TE/mg protein sample) was calculated by dividing the slope of the linear regression fit of each sample by the slope of the trolox linear regression fit for the same assay according to Eq ACE-inhibition activity The ACE-inhibition activity was as described by Shalaby, Zakora, and Otte (26) and involved taking, 1 ll of ACE solution (.25 units/ml in deionised water) and 3 ll of hydrolysate solution ( mg/ml in 5 mm Tris HCl buffer, ph 7.5, containing 3 mm NaCl) and placed separately in each well of a 96 well microtiter plate. Prior to the analysis, the hydrolysate solutions were centrifuged at 55g for 5 min to remove particulate ð1þ ð2þ ð3þ ð4þ matter. The plates were incubated at 37 C for 5 min in a microplate reader (FLUOstar Omega plate reader; BMG, GmbH, Offenburg, Germany), and after incubation, the plate reader was programmed to dispense 15 ll of FAPGG (.88 mm in the same buffer) into each well after the first cycle. The control solution was a 3 ll of the Tris HCl buffer and was included in each batch. All analyses were performed in duplicate with each replicate analysed in two independent runs made for each sample. The decay in the absorbance at 34 nm due to the degradation of FAPGG by ACE and the inhibitory properties of the hydrolysates were monitored for 5 min. The ACE activity was expressed as the slope of the decrease in the absorbance at 34 nm (DA) over a linear interval from the 1th to the 35th min. Within this time interval, absorbance signals were stable with little indications of signal drifting. The ACEinhibition activity (%) of the WPI hydrolysates was calculated as: % ACE inhibition ¼ 1 DA ðinhibitorþ DA ðcontrolþ 1 where DA (inhibitor) and DA (control) are the slopes of the samples with hydrolysate and of the control, respectively Surface hydrophobicity (S o ) The surface hydrophobicity was performed as reported by Alizadeh-Pasdar and Li-Chan (2) and used the anionic fluorescence probe 1-anililo-naphthalene-8-sulfonate (ANS). The protein concentration ranged from.5% to.25% (g/ml) in 1 mm phosphate buffer (ph 7), while the ANS concentration was maintained at 8 mm. The excitation wavelength was 39 nm while the emission wavelength was set at 47 nm. The relative fluorescence intensity (RFI) was measure using a Cary Eclipse Fluorescence Spectrophometer (Varian, Inc., Victoria, Australia). The S o was determined from the initial slope of the linear regression analysis of the plot of RFI against protein concentration (%). Each sample was run in triplicate Statistical analysis The entire study was done in duplicate or triplicate and statistical analysis was performed using a one-way ANOVA. The relationship between the three experimental factors was tested using Fisher s least significant difference (LSD) test. Multiple range comparison tests were further used to determine whether sample means were statistically different and a t-test was used to determine statistically significant difference among two groups (P <.5). ANOVA and t-tests were obtained using Statgraphics Ò Centurion XVI (StatPoint Technologies, Inc., Warrenton, VA, USA). 3. Results and discussion 3.1. Degree of hydrolysis The number of peptide bonds cleaved is quantified as the DH. The DH increased with hydrolysis time and the maximum DH reached for all samples ranged from 11.8 ±.2% to 14.1 ± 1.7% (Fig. 1). With the exception of chymotrypsin, heat pre-treatment marginally increased the DH for all enzymes. The order of DH for the heat pre-treated WPI was pepsin > trypsin > chymotrypsin for the 12-h hydrolysis and pepsin > chymotrypsin > trypsin for the 24-h hydrolysis. For the non heat pre-treated WPI, the DH order was chymotrypsin > pepsin > trypsin and was independent of hydrolysis time. Hydrolysates resulting from the three enzymes showed no significant difference in DH within the same pre-treatment condition and hydrolysis time (P >.5) and likewise, no significant difference was shown between the 12- and 24-h ð5þ

4 1438 R. Adjonu et al. / Food Chemistry 136 (213) Hours 24 Hours 1 A B Degree of hydrolysis (%) N-C H-C N-P H-P N-T H-T Pre-treatment condition and enzyme type Fig. 1. Degree of hydrolysis (DH) for WPI hydrolysates after 12 and 24 h of hydrolysis. N-C = Non heat pre-treated WPI hydrolysed with chymotrypsin, H- C = Heat pre-treated WPI hydrolysed with chymotrypsin; N-P = Non heat pretreated WPI hydrolysed with pepsin, H-P = Heat pre-treated WPI hydrolysed with pepsin; N-T = Non heat pre-treated WPI hydrolysed with trypsin, H-T = Heat pretreated WPI hydrolysed with trypsin. Absorbance at 214 nm /mau (Arbitrary units) C E D F hydrolysis time for the same pre-treatment condition and enzyme type (P >.5). While the DH was consistently higher in heat pretreated samples for the peptic and tryptic digests (Fig. 1), there was no statistically significant difference (P >.5). Normally, different enzymes will hydrolyse proteins to different degrees due to the high enzyme-peptide bond specificity. Both chymotrypsin and trypsin are alkaline endoproteases hydrolysing peptide bonds at the C-terminus of aromatic amino acids (tryptophan, tyrosine, and phenylalanine) and at the C-terminus of arginine and lysine residues, respectively. Pepsin on the other hand, is an acidic endoprotease hydrolysing peptide bonds at the N-terminus of aromatic and hydrophobic amino acids. The relatively high DH with pepsinolysis in the heat pre-treated WPI can be partly attributed to conformational changes in the WPI with heat pre-treatment as well as structural changes resulting from the working ph of 2.6. Pihlanto-Leppälä et al. (2), reported a similar trend, where heat pre-treated b-lactoglobulin was easily hydrolysed by pepsin and had a high DH compared to a non heat pre-treated sample. Conformational changes due to heat denaturation can result in a re-orientation of the peptide bonds and the exposure of previously hindered cleavage sites. These structural changes can lead to improved hydrolytic cleavage or the masking of certain previously exposed sites to enzymatic action, as evidenced in the chymotrypsin hydrolysis in this study. The low DH obtained after 24 h of hydrolysis may be due to the globular nature of whey proteins, which can limit enzymatic interactions. Peng, Xiong, and Kong (29) obtained a DH of about 35.% for WPI treated with Alcalase after 5 h of hydrolysis. The DH is high compared to the present study, possibly due to the broad specificity of Alcalase (a non-specific protease, though favouring aromatic or acidic amino acids) compared to that of the digestive enzymes used in this study. Additionally, the differences in the DH between the two studies may be due to the method used to obtain the DH. Peng et al. (29) used the ph stat method in their study which has been shown to give higher values for DH compared to the OPA method that was used in this study (Spellman, McEvoy, O Cuinn, & FitzGerald, 23), especially in cysteine rich proteins like WPI. This was attributed to cysteine and OPA forming isoindole, which is unstable and tends to underestimate the DH. The DH results presented here are comparable to those of Cheison et al. (21), where a DH of only 9.% was obtained after hydrolysis of b-lg by trypsin for 26 h. In part, they attributed the low DH to the globular nature of the whey protein which makes it difficult for high DHs to be obtained Retention time (minutes) Fig. 2. Size exclusion profile of WPI hydrolysates after hydrolysis for 12 or 24 h. A = Unhydrolysed WPI (Non heat pre-treated), B = Unhydrolysed WPI (Heat-pre treated), C = Chymotrypsin (Non heat pre-treated and heat pre-treated), D = Pepsin (Non heat pre-treated and heat pre-treated), E = Trypsin (Non heat pre-treated) and F = Trypsin (Heat pre-treated) Peptide profile and molecular weight (MW) distribution Fig. 2 shows the chromatograms (peptide profile) of WPI hydrolysates detected at 214 nm and the molecular weight of the peptides observed after gel filtration chromatography is shown in Table 1. Generally, hydrolysis time had no observable effect on the peptide profile for the same enzyme, though there were differences in the peptide profiles between enzymes at the same hydrolysis time. While heat pre-treatment changed the observed profile for the trypsin hydrolysates, no difference was apparent for the chymotrypsin and pepsin hydrolysates. The peptide profile highlights the effect of enzyme type (specificity), indicating that there are differences in the characteristic peptide fragments produced during hydrolysis. Although there was no observable differences in the size exclusion chromatograms between the heat pre-treated and non heat pre-treated samples for the chymotrypsin and pepsin hydrolysates, this does not rule out possible differences in the composition of the peptides and hence, their characteristic functionalities (Lourenco da Costa et al., 27). Only two peaks were detected in the chymotrypsin hydrolysates (Table 1). Chymotrypsin hydrolysis of non heat pre-treated WPI tended to produce smaller MW peptides (41%) compared to the heat pre-treated WPI (36%), which is consistent with the DH values obtained in the previous section. The DH of non heat pretreated WPI was higher but not significant (P >.5) compared to the heat pre-treated WPI for both the 12 and 24 h of hydrolysis. The consistent results obtained by the size exclusion chromatography and DH method indicate that heat pre-treatment does affect the hydrolysis pattern and hence the size and relative proportions of peptides formed. The chromatogram for the peptic digest showed only a single peak which accounted for over 78% of the total chromatogram

5 R. Adjonu et al. / Food Chemistry 136 (213) Table 1 Molecular weight (MW) range and distribution (%) of detected peptides of WPI and its hydrolysates. Pre-treatment Non heat pre-treatment Heat pre-treatment Hydrolysis time 12 h 24 h 12 h 24 h Enzyme type MW (kda) % MW (kda) % MW (kda) % MW (kda) % Chymotrypsin Pepsin Trypsin Unhydrolysed WPI IC 5 (µg hydrolysate/ml) Non heat pre-treated Heat pre-treated a, k a, u Chymotrypsin Pepsin Enzyme type Trypsin (Table 1). There was a striking effect of hydrolysis time when the 12- and 24-h non heat pre-treated samples were examined. At 12 h the only peak detected accounted for 93% of the chromatogram, while at 24 h the only peak accounted for 78%. This may be due to extended hydrolysis (24 h) degrading peptides to form more tripeptides, dipeptides and free amino acids which probably were not detected. Also, the heat pre-treated WPI hydrolysates had higher DHs compared to the non heat pre-treated WPI hydrolysates at both hydrolysis times (Fig. 1). Heat pre-treatment enhanced pepsinolysis of WPI and as a result, may influence the characteristic nature of peptides released during enzymatic hydrolysis of food proteins. The major fraction detected for the tryptic hydrolysates accounted for over 8% of the whole chromatogram. The influence of heat pre-treatment was demonstrated by the 214 nm absorbance data, with two separate peaks for the non heat pre-treated sample (Table 1). For the non heat pre-treated WPI, an additional peak with MW between 56 and 85 kda was observed on the chromatogram which was absent in the heat pre-treated WPI (Fig. 3). This peak was absent in the other hydrolysates and was not detected in the native WPI profiles (Fig. 3A and B). Although the percentage abundance was relatively small (5 6%), its presence is considered important. Since trypsin hydrolyses arginine and lysine residues and these amino acids usually lie on the outer surface of a typical disordered polypeptide chain, residues buried in the rigid structures of globular proteins such as WPI may not be accessible to tryptic attack (Agboola & Dalgleish, 1996). Heat pretreatment may have resulted in the unfolding of the globular b, l a, v a, m a, v Fig. 3. ACE-inhibition activity of WPI hydrolysates measured as the IC 5 value. Different letters show means that are statistically significant from each other (a, b compares means between non heat pre-treated and heat pre-treated WPI for the same enzyme type; k, l, m, and u, v compare means between non heat pre-treated and heat pre-treated WPI, respectively). Table 2 Free radical scavenging activity (FRSA) of WPI and its hydrolysates. Pre-treatment Activity (lmol TE/mg protein sample) Non heat pre-treatment Heat pre-treatment Hydrolysis time 12 h 24 h 12 h 24 h Chymotrypsin.27 ±.4 a.31 ±.3 a.31 ±.2 a.28 ±.4 a Pepsin.32 ±.3 a.28 ±.3 a.3 ±.3 a.28 ±.1 a Trypsin.32 ±.1 a.3 ±.5 a.31 ±.2 a.26 ±.2 a Unhydrolysed WPI.8 ±.2 b.7 ±.4 b All data were expressed as means of triplicate measurements ± standard deviation (SD). ab Different letters signifies means which are significantly different from each other at P <.5. structure allowing additional hydrolysis to occur. Additionally, the presence of this chromatographic peak may be due to aggregates of immunoglobulins ( pentamers ), serum albumins and hydrolysed peptides which may have formed through hydrophobic interactions, during the post-hydrolysis heat treatment to inactivate the enzyme (Peña-Ramos, Xiong, & Arteaga, 24). The MW distributions of the all hydrolysates were however, consistent with that reported by Lourenco da Costa et al. (27), Peng et al. (29) and Pihlanto-Leppälä et al. (2). Another peak common to both the tryptic digests (3 5%) may have resulted from undegraded a-lactalbumin, b-lactoglobulin or bovine serum albumin which were absent in the chymotrypsin and pepsin digests. This suggests that most of the bonds available for enzymatic action were those formed by the aromatic amino acids tyrosine, tryptophan and phenylalanine because they are the specific bonds targeted for hydrolysis by chymotrypsin and pepsin. Thus, enzymatic hydrolysis may not only cleave peptide bonds, but could also result in the unmasking of certain concealed peptide sequences in the native structure of globular proteins Biological activities of WPI hydrolysates Antioxidant activities Free radical scavenging activity (FRSA). The FRSA of the hydrolysates measured as the trolox equivalence is shown in Table 2. No statistically significant differences were found in the antioxidant activities between hydrolysates due to pre-treatment, enzyme type and hydrolysis time (P >.5). The 12-h hydrolysates from heat pre-treated WPI had higher scavenging abilities on ABTS + compared to the 24-h hydrolysates for all enzyme types. The same trend was observed for the non heat pre-treated samples except for the chymotrypsin hydrolysates. While the 24-h non heat pre-treated chymotrypsin hydrolysates showed a higher absolute FRSA value, the difference between heat pre-treated and non heat pre-treated samples was not significant (P >.5). All of the

6 144 R. Adjonu et al. / Food Chemistry 136 (213) Table 3 Oxygen radical absorbance capacity (ORAC) of WPI and its hydrolysates. Pre-treatment ORAC value (lmol TE/mg protein sample) Hydrolysis time Non heat pre-treatment Heat pre-treatment 12 h 24 h 12 h 24 h Chymotrypsin.78 ±.2 aku.74 ±.3 alu.79 ±.4 aku.75 ±.11 aku Pepsin.62 ±.3 blu.65 ±.7 bmu.71 ±.5 aku.73 ±.3 aku Trypsin.67 ±.2 bmv.81 ±.2 aku.75 ±.2 aku.8 ±.4 aku Unhydrolysed WPI.23 ±.2 c.12 ±.6 d All data are means of triplicate determinations ± standard deviation (SD). Data for unhydrolysed WPI was not included in the analysis of variance. ab Different letters for the same enzyme type and hydrolysis time (i.e. in the same row) indicate significant differences (P <.5) in the effect of heat pre-treatment on ORAC values. cd Different letters indicate significant differences (P <.5: t-test) in ORAC values between non heat pre-treated and heat pre-treated unhydrolysed WPI. klm Different letters for the same pre-treatment condition and hydrolysis time (i.e. in the same column) indicate significant differences (P <.5) in the effect of enzyme type on ORAC values. uv Different letters for the same enzyme type and pre-treatment condition (i.e. in the same row) indicate significant differences (P <.5) in the effect of hydrolysis time on ORAC values. enzyme treated samples however, had significantly higher FRSA values than the unhydrolysed WPI (P <.5). This indicates that hydrolysis is essential for the liberation of antioxidant peptides Oxygen radical absorbance capacity (ORAC). The ORAC values of the hydrolysates are as shown in Table 3. Analysis of variance indicated significant differences between sample means (P <.5) for the different enzyme types, pre-treatment condition and hydrolysis times. Multiple range comparison tests or the t-test indicated significant differences were primarily dictated by the pre-treatment condition and the type of enzyme used (Table 3). Furthermore, with the exception of the tryptic hydrolysates of non heat pre-treated WPI (P <.5), hydrolysis time had no significant effect on the antioxidant activity for the other five hydrolysates (P >.5). The ORAC values varied from.71 to.8 (heat pre-treated) and.62 to.81 (non heat pre-treated) for both hydrolysis times. Based on the ORAC values, the three enzymes tested were ranked chymotrypsin > trypsin > pepsin (12-h hydrolysis) and trypsin > chymotrypsin > pepsin (24-h hydrolysis), irrespective of pre-treatment condition. The ORAC values obtained in this work were larger than that of an Alcalase-hydrolysed crude egg hydrolysate (.463 lmol TE/mg protein) reported by You, Udenigwe, Aluko, and Wu (21), but compares with that reported by Hernández-Ledesma et al. (25), especially for the peptic and tryptic hydrolysates but not for the chymotryptic hydrolysates. This may be explained by the fact that their work used a-lactalbumin and b-lactoglobulin whereas the work reported here used WPI. Consequently, the heterogeneous mixture of proteins in WPI (a-lactalbumin, b-lactoglobulin, bovine serum albumin and immunoglobulin) may affect the hydrolysis process and hence the nature of the peptides released. However, unlike the FRSA (ABTS), the 24-h hydrolysates showed consistently higher ORAC values compared to the 12-h hydrolysates with the exception of the chymotryptic hydrolysates. The ORAC and the FRSA values for the chymotryptic hydrolysates suggest that, prolonged hydrolysis may not be suitable for generating antioxidant peptides from chymotrypsin hydrolysis of heat pre-treated WPI, since prolonged hydrolysis may generate high proportions of free amino acids. At very high concentrations, these free amino acids can negatively affect the antioxidant activity by acting as prooxidants which may decrease the observed activity (Pihlanto, 26). Table 4 Correlations between FRSA and ORAC assays for hydrolysed and unhydrolysed WPI. Pre-treatment Correlation coefficients (R 2 ) Non heat pre-treatment Heat pre-treatment Hydrolysis time 12 h 24 h 12 h 24 h Chymotrypsin Pepsin Trypsin Unhydrolysed WPI In general, peptic hydrolysates of WPI were less effective in scavenging peroxyl radicals (ORAC) compared to the ABTS + free radicals (FRSA). Differences in antioxidant activity between assays are expected due to differences in the reaction mechanisms being measured, the variations in experimental techniques with respect to each antioxidant assay and the heterogeneity of the products of enzymatic hydrolysis. All hydrolysates in this work showed dose-dependency with the FRSA and ORAC assay which suggested that the reaction kinetics may not follow a 1:1 stoichiometry. Instead, reactions showed variable ratio stoichiometry which may be highly dose-dependent with respect to a particular antioxidant hydrolysate. Working over a concentration range of substrate antioxidant (WPI hydrolysates) to compute the net antioxidant activity will therefore be more accurate than the single concentrationpoint reading method. While the magnitude of the antioxidant activities of hydrolysates differed in FRSA and ORAC assays, the FRSA and ORAC values were positively correlated (R 2 =.72.99) for all hydrolysates, with lower correlations for unhydrolysed WPI samples (R 2 =.56) (Table 4). The relatively high correlation between the two methods suggests that, although their reaction pathways were different, the results complemented each other when it came to assessing the antioxidant activities of proteins and their hydrolysates, and may also provide some insight into their mechanisms of action. The low R 2 observed for the unhydrolysed WPI samples was probably due to the lower sensitivity of the unhydrolysed WPI samples towards the FRSA and the ORAC assays. Dudonné, Vitrac, Coutière, Woillez, and Mérillon (29) found correlations of between.62 and.85 for a variety of assays, including ORAC, FRSA (ABTS + ), superoxide dismutase (SOD), ferric reducing antioxidant potential (FRAP) and 2,2-diphenyl-1-picrylhydrazyl (DPPH), which is consistent with the results reported in the current study. Hydrolysis time had no significant effect on the DH or the size exclusion profiles for the hydrolysates, irrespective of the pretreatment condition and enzyme type used. Furthermore, Singh and Dalgleish (1998) reported that extensive hydrolysis (24 h) is detrimental to other functional properties of proteins (e.g. their emulsifying properties). As this study was aimed at generating dual functional peptides (i.e. exhibiting both biological and technological [nanoemulsification] properties) high proportions of free amino acids and oligopeptides may not be desirable for their emulsification abilities. It was apparent from the antioxidant activity data that extensive hydrolysis (24 h) did not improve the antioxidant activity, but in most cases, reduced the antioxidative potentials of the hydrolysates (Tables 2 and 3). Subsequent analyses for ACE-inhibition activity and surface hydrophobicity were therefore limited to 12 h hydrolysates ACE-inhibition activity The ACE-inhibition activity was expressed as the IC 5 value (i.e. the concentration of hydrolysate required to inhibit the activity of the ACE enzyme by 5%), and is as shown in Fig. 3. Based on the IC 5 values, hydrolysates from the three enzymes tested were ranked chymotrypsin > trypsin > pepsin for the non heat pre-treated WPI (P <.5) and chymotrypsin > pepsin > trypsin for the heat

7 R. Adjonu et al. / Food Chemistry 136 (213) pre-treated WPI (P <.5). Heat pre-treatment marginally enhanced the ACE-inhibition activity of the tryptic hydrolysates but not for the chymotryptic hydrolysates. The heat pre-treated peptic digest was about 5% more potent (IC 5 = 515 lg/ml; P <.5) compared to the non heat pre-treated digest (IC 5 = 131 lg/ml). This indicated that in addition to enhancing their hydrolysis, heat pre-treatment of WPI prior to its hydrolysis with pepsin is vital to their ACE-inhibitory properties. Chymotrypsin hydrolysates of WPI showed the lowest IC 5 value amongst the three enzymes irrespective of the pre-treatment condition used, indicating they have potential as dual functional peptides in the food industry (Tables 2 and 3). Structure activity studies have identified the contributions of hydrophobic amino acids (e.g. tryptophan, tyrosine, phenylalanine or proline) at the C-terminal to the ACE inhibition properties of peptides (Li et al., 24). The exposure of these aromatic hydrophobic amino acids at the C-terminal during the chymotrypsin hydrolysis of WPI may have contributed to their high inhibitory properties compared to the other hydrolysates. These C-terminal aromatic and hydrophobic residues is believed to influence the binding of the inhibitior to the S 1,S 1, and S 2 active site of ACE (Li et al., 24). Lourenco da Costa et al. (27), also reported the highest inhibitory activity by chymotrypsin hydrolysates of WPI compared to that obtained by other enzymes, such as Alcalase and Protamex. The greater inhibitory activity of the peptic digest of heat pretreated WPI compared to the non-heat pre-treated WPI may be attributed to changes in the conformation of bioactive peptides during heat pre-treatment (Lopez-Fandino, Otte, & Van Camp, 26). Heat pre-treatment may have resulted in the exposure of more hydrophobic peptides as well as amino acids which have been shown to contribute to the ACE-inhibitory activities of peptides (Li et al., 24; Lourenco da Costa et al., 27). In addition, the presence of the aliphatic/basic amino acids lysine and arginine (target amino acids for trypsin) at the C-terminal has also been implicated in the ACE-inhibitory potency of peptide fractions (Lopez-Fandino et al., 26). Rao et al. (211) indicated that the proportion of hydrophobic and basic amino acid residues in the hydrolysates and their sequence in the peptide play a major role in their ACE-inhibitory properties. The IC 5 values reported in this work compared favourably with values previously reported by Lourenco da Costa et al. (27), Pihlanto-Leppälä et al. (2) and Shalaby et al. (26). Similarly, as the antioxidant activities of the crude WPI hydrolysates resulted from the actions of several peptide components, the ACE-inhibition activity resulted from the synergistic effects of several peptides making up the crude hydrolysates. Moreover, due to the fact that there are different conformational requirements for the catalytic sites of ACE (Lopez-Fandino et al., 26) and coupled with the heterogeneity of the products of enzymatic hydrolysis, there exist a diverse range of potential peptides with varied conformational characteristics that can inhibit the activities of the ACE enzyme. These diverse peptides have, however, been reviewed by several authors (Korhonen & Pihlanto, 26; Lopez- Fandino et al., 26). The unhydrolysed WPI (heat pre-treated and non heat pre-treated) in this work only showed any ACE-inhibitory activity when used at approximately 1-fold the concentration of the hydrolysates, however, their IC 5 values could not be determined. Rao et al. (211), have also reported no ACE-inhibitory activity for an undigested hen egg white lysozyme, therefore underpinning the importance of hydrolysis in order to exploit the bioactive properties of WPI Surface hydrophobicity (S o ) The S o values ranged from to 89.3 and to 888. for the non heat pre-treated and heat pre-treated WPI samples, Surface hydrophobicity (S o ) Non heat pre-treated WPI Heat pre-treated WPI a, k a, m b, w b, x Chymotrypsin Pepsin Trypsin Unhydrolysed WPI Enzyme type Fig. 4. Surface hydrophobicity value of hydrolysed and unhydrolysed WPI. Different letters show means that are statistically significant from each other (a, b compares means between non heat pre-treated and heat pre-treated WPI for the same enzyme type; k, l, m, n and u, v, w, x compare means between non heat pretreated and heat pre-treated WPI, respectively). respectively (Fig. 4). The order for the S o was pepsin > unhydrolysed WPI > chymotrypsin > trypsin for non heat pre-treated, and trypsin > unhydrolysed WPI > chymotrypsin > pepsin for the heat pre-treated WPI. Heat pre-treatment significantly decreased the S o for the unhydrolysed WPI (P <.5) and for the chymotrypsin and pepsin hydrolysates, but significantly increased the S o for the trypsin hydrolysate (P <.5). The tryptic hydrolysate for the heat pre-treated WPI showed the highest S o value whereas the peptic hydrolysate showed the least, and vice versa for the non heat pre-treated WPI. The magnitude of the S o values compared favourably with those reported by Alizadeh-Pasdar and Li-Chan (2) and Lee et al. (1992). The decrease in the S o of the unhydrolysed heat pre-treated WPI was consistent with the observations of Mutilangi, Panyam, and Kilara (1996) but not those of Alizadeh-Pasdar and Li-Chan (2). The are two possible reasons for this observation; the first is that interactions of sulphydryl groups as well as intermolecular hydrophobic interactions during heat denaturation of WPI may act to reduce the number of exposed hydrophobic residues, resulting in a reduction in the S o (Mutilangi et al., 1996). The second reason is that the S o of proteins may increase in the initial phase of thermal denaturation due to the relaxation of the protein structure, facilitating the access of ANS to previously hindered hydrophobic sites, however, as heating progresses, there is loss of ANS binding sites as a result of the shrinking and compactness of the protein molecule which may cause a reduction in S o (Eynard, Iametti, Relkin, & Bonomi, 1992). Both chymotrypsin and pepsin specifically hydrolyse peptide bonds formed by aromatic and hydrophobic amino acids. Heat pre-treatment of the WPI prior to hydrolysis may have resulted in exposure of more aromatic and hydrophobic residues. These bonds could be destroyed during hydrolysis accounting for the reduction in their S o values. The peptic hydrolysate from heat pre-treated WPI however, recorded the least S o value compared with the chymotryptic and tryptic hydrolysates. This may have resulted from the high DH obtained for the peptic hydrolysate. The S o value increased for DH values ranging from 2.8% to 4.3% and then decreased at DH of about 8.% (Mutilangi et al., 1996). The high DH (13.4%) obtained after the pepsin hydrolysis of heat pre-treated WPI for 12 h is indicative of a greater proportion of free amino acids and short chain peptides. Greater proportions of especially hydrophilic amino acids and peptides compared to that of hydrophobic residues could exert a fluorescence quenching-effect on b, n a, u a, l b, v

8 1442 R. Adjonu et al. / Food Chemistry 136 (213) the observed fluorescence of ANS probe. Competitive inhibitorybinding of hydrophilic residues to ANS probe may reduce the energy transfer from peptide bonds to the ANS probe and hence reduce the probes quantum yield of florescence (Haskard & Li-Chan, 1998). Hydrolysis of heat pre-treated WPI by trypsin increased significantly the S o value (P <.5). As discussed in Section 3.2, trypsin hydrolyses peptide bonds formed by arginine and lysine residues which are located on the outer surface of the polypeptide chain of globular proteins. Heat pre-treatment may have resulted in the unfolding of the globular structure and enhanced the hydrolysis by trypsin, and resulted in the exposure of more ANS specific binding sites, resulting in an increase in S o. In addition, since the hydrolysis by trypsin was not targeted at peptide bonds formed by the aromatic amino acids, these bonds are likely to be preserved in the trypsin hydrolysis of heat pre-treated WPI, allowing their interaction with the ANS probe. Generally, the S o of the hydrolysates were mostly reduced compared to the unhydrolysed WPI. This is because hydrolysis can result in the unfolding and subsequent disruption of the secondary and tertiary structures of protein molecules. A disruption of the organised hydrophobic sites recognisable by the ANS probe would result in a reduction in S o (Eynard et al., 1992). The effect of hydrolysis on S o is suggestive of the possible contributions of the secondary and tertiary structures of proteins to their hydrophobic properties and not just their hydrophobic amino acid residues. The reduction observed in this work was highly dependent on a combination of heat pre-treatment and enzyme type. Controlled hydrolysis to partially unfold and expose the hydrophobic core would therefore be beneficial in order to enhance the hydrophobic surface properties of WPI hydrolysates. In the present work, all hydrolysates showed varying S o values which is indicative of differences in the total number of hydrophobic groups generated by hydrolysis of WPI with the three enzyme types. Aside from the molecular weight (size) of peptides, an increase in the number of hydrophobic groups could facilitate interactions between proteins and oil at the oil/water interface during emulsion formation (Mahmoud, Malone, & Cordle, 1992). A correlation between the surface hydrophobicity of proteins and their emulsifying activity has been observed (Mahmoud et al., 1992). The more hydrophobic the protein (higher S o value), the greater the ability of the protein to decrease the interfacial tension between oil and water and to ultimately stabilise the oil droplets during emulsion formation. On the other hand, the antioxidative and ACE-inhibition activities of peptides have been associated with their hydrophobic properties (Lopez-Fandino et al., 26; Pihlanto, 26). Knowledge of the hydrophobic properties of WPI hydrolysates will give insight into the biological and technological functionalities of peptides, such as their interactions with oil droplets during emulsion formation (Haskard & Li-Chan, 1998). 4. Conclusion The influence of heat pre-treatment and the specificity of the hydrolysing enzyme on the antioxidant and ACE-inhibition activities of crude WPI hydrolysates were investigated. Heat pretreatment showed no apparent effect on the molecular weight distribution and antioxidant activity of the WPI hydrolysates, but had a significant impact on the ACE-inhibitory activity. Additionally, heat pre-treatment significantly affected the hydrophobic properties of the peptides, and this was highly dependent upon the specificity of the enzyme. The molecular weights of the peptides in the hydrolysates were highly varied, with the peptides MW ranging from about.3 to 1 kda. This indicated a wide range of sufficiently large peptides able to stabilise food nanoemulsion systems because this category of emulsions have particularly small droplet sizes. As this study was aimed at screening of the crude WPI hydrolysates for their dual functionalities, bioactivity and hydrophobicity results have shown that it is possible to exploit peptides from WPI hydrolysates for their dual functionalities in functional foods manufacture through a combination of heat pretreatment and careful enzyme selection. Of the three most common endopeptidases, chymotrypsin produced peptides with highest antioxidant and ACE-inhibitory activities, followed by trypsin and then pepsin. The chymotrypsin hydrolysates showed a broad MW size range followed by pepsin and then trypsin, which may explain the variations in the observed functionalities. Research on the role of specific peptide fractions obtained by ultrafiltration on the observed functionalities as well as their ability to stabilised food nanoemulsion systems is underway in order to elucidate their potential as dual functional ingredients in food systems. Acknowledgements The authors would like to acknowledge the support of the EH Graham Centre for Agriculture Innovation towards this research work. Author Adjonu is a recipient of the Charles Sturt University International Postgraduate Research Scholarship (IPRS). References Agboola, S. O., & Dalgleish, D. G. (1996). Enzymatic hydrolysis of milk proteins used for emulsion formation. 1. Kinetics of protein breakdown and storage stability of the emulsions. Journal of Agricultural & Food Chemistry, 44(11), Alizadeh-Pasdar, N., & Li-Chan, E. C. Y. (2). Comparison of protein surface hydrophobicity measured at various ph values using three different fluorescent probes. Journal of Agricultural & Food Chemistry, 48(2), Cheison, S. C., Schmitt, M., Leeb, E., Letzel, T., & Kulozik, U. (21). Influence of temperature and degree of hydrolysis on the peptide composition of trypsin hydrolysates of b-lactoglobulin: Analysis by LC-ESI-TOF/MS. Food Chemistry, 121(2), Custódio, M. F., Goulart, A. J., Marques, D. P., Giordano, R., Giordano, R. L. C., & Monti, R. (29). Hydolysis of cheese whey proteins with trypsin, chymotrypsin and carboxypeptidase A. Alimentos e Nutrição Araraquara, 16(2), Dissanayake, M., & Vasiljevic, T. (29). Functional properties of whey proteins affected by heat treatment and hydrodynamic high-pressure shearing. Journal of Dairy Science, 92(4), Dudonné, S., Vitrac, X., Coutière, P., Woillez, M., & Mérillon, J. M. (29). Comparative study of antioxidant properties and total phenolic content of 3 plant extracts of industrial interest using DPPH, ABTS, FRAP, SOD, and ORAC assays. Journal of Agricultural & Food Chemistry, 57(5), Eynard, L., Iametti, S., Relkin, P., & Bonomi, F. (1992). Surface hydrophobicity changes and heat-induced modifications of a-lactalbumin. Journal of Agricultural & Food Chemistry, 4(1), Gauthier, S., & Pouliot, Y. (23). Functional and biological properties of peptides obtained by enzymatic hydrolysis of whey proteins. Journal of Dairy Science, 86, E78 E87. Haskard, C. A., & Li-Chan, E. C. Y. (1998). Hydrophobicity of bovine serum albumin and ovalbumin determined using uncharged (PRODAN) and Anionic (ANS ) fluorescent probes. Journal of Agricultural & Food Chemistry, 46(7), Hernández-Ledesma, B., Davalos, A., Bartolome, B., & Amigo, L. (25). Preparation of antioxidant enzymatic hydrolysates from a-lactalbumin and b-lactoglobulin. Identification of active peptides by HPLC-MS/MS. Journal of Agricultural & Food Chemistry, 53(3), Korhonen, H., & Pihlanto, A. (26). Bioactive peptides: Production and functionality. International Dairy Journal, 16(9), Lee, S.-Y., Morr, C. V., & Ha, E. Y. W. (1992). Structural and functional properties of caseinate and whey protein isolate as affected by temperature and ph. Journal of Food Science, 57(5), Lee, Y., Skurk, T., Hennig, M., & Hauner, H. (27). Effect of a milk drink supplemented with whey peptides on blood pressure in patients with mild hypertension. European Journal of Nutrition, 46(1), Li, G. H., Le, G. W., Shi, Y. H., & Shrestha, S. (24). Angiotensin I-converting enzyme inhibitory peptides derived from food proteins and their physiological and pharmacological effects. Nutrition Research, 24(7), Lopez-Fandino, R., Otte, J., & Van Camp, J. (26). Physiological, chemical and technological aspects of milk-protein-derived peptides with antihypertensive and ACE-inhibitory activity. International Dairy Journal, 16(11), Lourenco da Costa, E., Antonio da Rocha Gontijo, J., & Netto, F. M. (27). Effect of heat and enzymatic treatment on the antihypertensive activity of whey protein hydrolysates. International Dairy Journal, 17(6), Madureira, A., Tavares, T., Gomes, A. M. P., Pintado, M., & Malcata, F. X. (21). Invited review: Physiological properties of bioactive peptides obtained from whey proteins. Journal of Dairy Science, 93(2),