See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/271539564

Effects of exopeptidase treatment on

antihypertensive activity and taste attributes of
enzymatic whey protein hydrolysates

Article in Journal of Functional Foods March 2015

Impact Factor: 3.57 DOI: 10.1016/j.jff.2014.12.036

CITATIONS READS

7 152

4 authors, including:

Rotimi Aluko Margaret Anne Cliff

University of Manitoba Agriculture and Agri-Food Canada
138 PUBLICATIONS 3,310 CITATIONS 134 PUBLICATIONS 2,081 CITATIONS

SEE PROFILE SEE PROFILE

All in-text references underlined in blue are linked to publications on ResearchGate, Available from: Margaret Anne Cliff
letting you access and read them immediately. Retrieved on: 22 May 2016
 journal of functional foods 13 (2015) 262275

Available online at www.sciencedirect.com

ScienceDirect

j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / j ff

Effects of exopeptidase treatment on

antihypertensive activity and taste attributes of
enzymatic whey protein hydrolysates

Lennie K.Y. Cheung a, Rotimi E. Aluko b, Margaret A. Cliff c,

Eunice C.Y. Li-Chan a,*
a
Food, Nutrition & Health Program, Faculty of Land & Food Systems, The University of British Columbia, 2205
East Mall, Vancouver, BC V6T 1Z4, Canada
b
Department of Human Nutritional Sciences and the Richardson Centre for Functional Foods and
Nutraceuticals, University of Manitoba, Winnipeg, MB R3T 2N2, Canada
c
Agriculture and Agri-Food Canada, Pacific Agri-Food Research Centre, P.O. Box 5000, Summerland, BC V0H
1Z0, Canada

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

Article history: The objectives of this study were to investigate the effects of exopeptidase treatment on
Received 22 June 2014 ACE-inhibitory activity, antihypertensive activity and taste of whey protein hydrolysates (WPHs).
Received in revised form 14 WPH with high ACE-inhibitory activity (IC50 = 0.15 mg/mL) was treated with carboxypepti-
December 2014 dase (Accelerzyme CPG), aminopeptidase (Peptidase R), or an aminopeptidase and proteinase
Accepted 19 December 2014 mixture (ProteAX). The exopeptidase-treated hydrolysates exhibited ACE-inhibitory activ-
Available online ity (IC50 = 0.240.34 mg/mL) and decreased systolic blood pressure (12 to 31 mm Hg) in
spontaneously hypertensive rats for 24 h after a single administration of 100 mg/kg body
Keywords: weight. The highest ACE-inhibitory activity was associated with the 2001000 Da fractions
Whey protein hydrolysates in all exopeptidase-treated hydrolysates. Exopeptidase treatment significantly lowered bit-
Bioactive peptides terness, increased umami and salty tastes, and increased overall acceptability of the starting
Enzymatic debittering WPH, changes that may be due to release of certain terminal amino acids. Therefore, exo-
Angiotensin I-converting enzyme peptidase treatment may be a viable debittering method for bitter-tasting, antihypertensive
inhibition protein hydrolysates, before incorporation into functional foods.
Antihypertensive activity 2014 Elsevier Ltd. All rights reserved.
Sensory evaluation

inhibitors are the class of drugs most commonly prescribed

1. Introduction
Hypertension, the main risk factor for cardiovascular disease,
accounts for 45 and 51% of heart disease- and stroke-related
deaths worldwide, respectively, and causes more than nine
million deaths annually (World Health Organization, 2013). In
Canada, synthetic angiotensin I-converting enzyme (ACE)
for cardiovascular diseases (Public Health Agency of Canada,
2009). These drugs inhibit ACE, an enzyme that promotes
sodium retention and vasoconstriction, to effectively control
hypertension (Ritter, 2011). Although effective, synthetic ACE
inhibitors are associated with undesirable side effects such as
chronic dry cough or angioedema, the latter of which can lead
to serious respiratory distress requiring ventilatory support or,

* Corresponding author. Food, Nutrition & Health Program, Faculty of Land & Food Systems, The University of British Columbia, 2205 East
Mall, Vancouver, BC V6T 1Z4, Canada. Tel.: +1 604 822 6182; fax: +1 604 822 5143.
E-mail address: eunice.li-chan@ubc.ca (E.C.Y. Li-Chan).
http://dx.doi.org/10.1016/j.jff.2014.12.036
1756-4646/ 2014 Elsevier Ltd. All rights reserved.
 journal of functional foods 13 (2015) 262275
in rare cases, death (Weber & Messerli, 2008). Therefore, there
is growing interest in natural ACE inhibitors in the form of pep-
tides from food protein hydrolysates, which have not been
associated with the aforementioned side effects, for use in func-
tional foods aimed to help in the management of hypertension.
However, bitterness has been experimentally correlated with
ACE-inhibitory activity in protein hydrolysates (Cheung &
Li-Chan, 2010) and predicted by computer modeling of pep-
tides, particularly those of shorter chain lengths (Pripp & Ard,
2007; Tan et al., 2013). Furthermore, the most bitter fraction
of a whey protein hydrolysate also had the highest ACE-
inhibitory activity (Cheison, Wang, & Xu, 2007). This relationship
between ACE inhibition and bitterness may exist because both
attributes have been associated with low molecular weight pep-
tides containing hydrophobic amino acid residues. Hydrophobic
amino acid residues, especially at the C-terminal, have been
shown to be highly influential of ACE-inhibitory activity in pep-
tides (Sagardia, Roa-Ureta, & Bald, 2013; Wang et al., 2011; Wu,
Aluko, & Nakai, 2006a, 2006b). Fractions with molecular
weights <1000 Da have also been shown to have higher ACE-
inhibitory activity than higher molecular weight fractions
(OLoughlin, Murray, FitzGerald, Brodkorb, & Kelly, 2014;
Pihlanto-Leppl, Koskinen, Piilola, Tupasela, & Korhonen, 2000).
Similarly, hydrophobic amino acid residues, particularly at the
C-terminal, and molecular weights <1000 Da have been corre-
lated with peptide bitterness using computer modeling (Kim
& Li-Chan, 2006).
Since hydrolysates often taste bitter, an attribute that can
decrease acceptance and even lead to the rejection of foods
(Temussi, 2011), the use of hydrolysates as functional food
ingredients can risk decreased consumer acceptance of the
target food product. Consumers are reluctant to compromise
poor taste for health benefits in functional foods (Verbeke,
2006), with factors such as flavor and texture influencing the
continued consumption of new food products regardless of
their health benefits (Barrios, Bayarri, Carbonell, Izquierdo, &
Costell, 2008). Various methods have been explored for
debittering protein hydrolysates to increase their palatabil-
ity, such as the removal of hydrophobic peptides, masking of
bitter taste, and encapsulation of hydrolysates (Hernndez-
Ledesma, Contreras, & Recio, 2011). Treatment with
exopeptidases, which are enzymes that release terminal amino
acid residues from peptides, is another promising debittering
method that reduces the bitterness of hydrolysates without
decreasing yields or requiring the addition of taste-active
compounds (Saha & Hayashi, 2001). Aminopeptidases and
carboxypeptidases are two types of exopeptidases that cleave
amino acid residues from the N- and C-terminal of peptides,
respectively. Exopeptidase treatment can decrease hydroly-
sate bitterness by releasing hydrophobic amino acids from
the N-terminal (Nishiwaki, Yoshimizu, Furuta, & Hayashi,
2002), C-terminal (Fu, Li, & Yang, 2011), or both terminals (Ge
& Zhang, 1996). Although effective for debittering, however,
hydrolysis with exopeptidases may result in cleaving off amino
acid residues pertinent to ACE-inhibitory activity. To date,
there is limited knowledge on the effects of exopeptidase
treatment on ACE-inhibitory activity, or any other bioactivity,
of protein hydrolysates. The effect of exopeptidase treatment
on the full taste profile of protein hydrolysates has also been
rarely addressed.
263
The objective of this study was to assess whether exopep-
tidase treatment is an appropriate debittering method for
hydrolysates with ACE-inhibitory activity. Hydrolysates of whey
protein have been widely studied for ACE-inhibitory activity
(Abubakar, Saito, Kitazawa, Kawai, & Itoh, 1998; OLoughlin et al.,
2014; Pihlanto-Leppl et al., 2000; Wang, Mao, Cheng, Xiong,
& Ren, 2010; Wang et al., 2012) and have been reported to have
bitter taste (Cheison et al., 2007; Leksrisompong, Gerard,
Lopetcharat, & Drake, 2012; Leksrisompong, Miracle, & Drake,
2010). Therefore, whey protein was selected as the substrate
for producing hydrolysates in the current study. The effects of
exopeptidase treatment on in vitro ACE-inhibitory activity, mo-
lecular size distribution, in vivo antihypertensive activity in
spontaneously hypertensive rats, as well as taste, overall ac-
ceptability and amino acid profile of the resulting whey protein
hydrolysates were investigated.
2. Materials and methods
2.1. Materials
NZMP whey protein isolate (WPI) 895 donated by Fonterra
Co-operative Group (Rosemont, IL) was used as the protein
source for hydrolysates production. Protease M Amano SD,
Protease P Amano 6SD, Thermoase PC10F, Protin SD-AY10,
Protin SD-NY10, Peptidase R, and ProteAX were donated by
Amano Enzyme U.S.A, Ltd. (Elgin, IL). Validase Papain Liquid,
Maxazyme NNP DS, and Accelerzyme CPG were donated by
DSM Food Specialties B.V. (Delft, The Netherlands). WPH 4003
was donated by PGP International (Eagan, MN), Hilmar 8350
and Hilmar 8390 were donated by Hilmar Ingredients (Hilmar,
CA), and NZMP WPH 917 (Fonterra Co-operative Group) was
donated by Caldic Canada Inc. (Delta, BC).
Trichloroacetic acid (TCA), food grade hydrochloric acid (HCl),
sodium hydroxide (NaOH) and caffeine meeting Food Chemi-
cal Codex requirements, and citric acid anhydrous meeting
United States Pharmacopeia requirements were from Fisher
Scientific (Fairlawn, NJ). Hippuryl-His-Leu-OH (HHL) was from
New Bachem (Torrance, CA), while 2,4,6-trinitrobenzenesulfonic
acid (TNBS) was from Thermo Scientific (Rockford, IL). Rabbit
lung angiotensin I-converting enzyme, spectrophotometric-
grade ethyl acetate, blue dextran, vitamin B12, l-carnosine, and
Leu were from Sigma-Aldrich (St. Louis, MO). An antifreeze
protein was donated by A/F Protein Canada Inc. (St. Johns, NF).
Trp and Ala were from Bio Basic Inc. (Markham, ON). All other
laboratory chemicals were of reagent grade. Monosodium glu-
tamate was from Ajinomoto North America, Inc. (Teaneck, NJ).
Iodized table salt (Windsor; Anjou, QC) and sugar (Rogers Sugar
Inc.; Vancouver, BC) were purchased locally.
2.2. Hydrolysate production
Hydrolysates were produced in the University of British Co-
lumbia (UBC) Food Science program pilot plant (Vancouver,
BC) following safe food handling practices. WPI was dis-
solved in distilled-deionized water (ddH 2 O) at 3 g WPI/
100 mL, resulting in a solution with pH 6.86 0.09, and pre-
heated with constant stirring in a temperature-controlled water
bath (Blue M Electric Company, Blue Island, IL). Incubation
264 journal of functional foods 13 (2015) 262275
temperatures were those specified for optimal enzyme activ-
ity by the manufacturers, as follows: Protease M Amano SD
(50 C), Protease P Amano 6SD (40 C), Thermoase PC10F (65 C),
Protin SD-AY10 (50 C), Protin SD-NY10 (50 C), Validase Papain
Liquid (65 C) and Maxazyme NNP DS (45 C). The
endoproteases (3 g/100 g WPI) were added to the pre-heated
WPI solution and incubated for a hydrolysis time of 3 h, which
was found in preliminary studies to be sufficient for releas-
ing ACE-inhibiting peptides. The sample codes for the resulting
WPHs were PM3, PP3, T3, AY3, NY3, VP3, and MX3, respec-
tively. After 3 h of incubation, aliquots (0.5 mL) were taken
for extent of hydrolysis analysis before hydrolysis was termi-
nated by heating to 90 C for 15 min. The mixture was then
centrifuged (10 min at 13,200 g) using a DuPont Sorvall Cen-
trifuge RC 5B (Mandel Scientific Co. Ltd., Guelph, ON) and the
supernatant lyophilized and stored at 20 C until use.
2.3. Exopeptidase treatment
The hydrolysate produced after a 3 h hydrolysis with Thermoase
PC10F (T3), as described in Section 2.2, was reconstituted to
10 g hydrolysate/100 mL in ddH2O and used as the starting ma-
terial for exopeptidase treatments. This sample was pre-
heated under constant stirring in a temperature-controlled
water bath until reaching incubation temperature, after which
it was treated with Accelerzyme CPG (AC), Peptidase R (PR),
or ProteAX (PX) at 4 g/100 g for 1 or 7 h. The incubation tem-
perature and pH used reflect those for optimal enzyme activity,
as provided by the suppliers. These conditions, as well as ad-
ditional characteristics of the exopeptidases, are detailed in
Table 1. Hydrolysis was terminated by heating to 90 C for
15 min and the hydrolysates lyophilized before storing at 20 C
until use. AC-treated hydrolysates were adjusted to pH 6 using
3 M NaOH prior to lyophilization. The hydrolysate treated with
AC, PR, or PX was abbreviated as T3-AC, T3-PR or T3-PX, re-
spectively. Sample codes were followed by a 1 or 7 to denote
1 or 7 h of exopeptidase treatment, respectively.
2.4. Extent of hydrolysis
The -amino content of protein hydrolysates, determined in
triplicate using the TNBS method described in Liceaga-Gesualdo
Table 1 Characteristics of the exopeptidases used for debittering a whey protein hydrolysate.
Product name Supplier description
(microbial source) (amino acid specificity)
Peptidase R (Rhizopus High peptidase activity for
oryzae) debittering bitter protein
hydrolysates (Arg > Ala, Lys,
Phe, Leu)
ProteAX (Aspergillus oryzae) High peptidase and
proteinase activity (Gln, Ser,
Thr, Met > Arg > Ala, Lys,
Phe, Leu)
Accelerzyme CPG Accelerates ripening and
(genetically modified promotes flavor in cheese
Aspergillus niger strain production (broad activity
ISO-528) with highest specificity for
Phe, Ile, Leu, Met, Val)
a
Information from FDA-GRN 000345 (U.S. Food and Drug Administration, 2010).
and Li-Chan (1999), was used to indicate the extent of hydro-
lysis (EH). Sample absorbance values at 420 nm were compared
to a Leu standard curve (03.0 mM Leu) constructed each day
the assay was conducted. The EH was reported as
milliequivalents of Leu (meq Leu)/g sample.
2.5. ACE-inhibitory activity
The protocol used for determining ACE-inhibitory activity was
as described in Cheung and Li-Chan (2010) with minor modi-
fications. Samples were reconstituted in 0.050 M sodium
tetraborate buffer with 0.3 M NaCl (pH 8.3) to 0.10.3 mg/mL
final assay concentrations. The released hippuric acid was ex-
tracted in ethyl acetate, which was evaporated until dry at
120 C. Samples with 5% ACE-inhibitory activity were con-
sidered not inhibitory. The IC50 value of hydrolysates (the
concentration required to inhibit 50% of the enzyme activity)
was obtained by cubic polynomial regression analysis of ACE-
inhibitory activities (%) at five or more sample concentrations
(final assay concentrations of 0.070.71 mg hydrolysate/mL).
2.6. Size exclusion chromatography
T3 and the 7 h exopeptidase-treated hydrolysates were recon-
stituted to 20 mg/mL in 0.05 M tetraborate buffer (pH 8.3) and
filtered through a 0.45 m filter (MILLEXHV filter unit, Millipore,
Cork, Ireland). The sample (500 L) was loaded into a fast protein
liquid chromatography system (FPLC) (KTA purifier 100/10; GE
Healthcare Life Sciences, Baie dUrfe, QC) and eluted through
the Superdex Peptide 10/300 GL column (24 mL, GE Health-
care, Uppsala, Sweden) at a flow rate of 0.2 mL/min using
tetraborate buffer as the eluent solvent. The fractionation range
of this column is 1007000 Da. Elution profiles were moni-
tored at 215 nm (mV) and 280 nm (Abs), and the eluate collected
in 0.4 mL volumes prior to pooling into fractions based on peaks
obtained in the elution profiles. The fractions were then lyo-
philized and stored at 20 C until use. After each elution, the
column was washed with ddH2O and equilibrated with buffer.
The molecular weight ranges of the fractions were estimated
based on the elution volumes of the following molecular weight
standards: antifreeze protein (3240 Da), vitamin B12 (1355.47 Da),
Enzyme class Hydrolysis conditions Manufacturer
3.4.11.1 leucyl amino- 52 C Amano Enzyme USA, Elgin,
peptidase pH 6 IL
3.4.11.1 leucyl amino- 60 C Amano Enzyme USA, Elgin,
peptidase pH 6 IL
3.4.16.x serine-type 37 C DSM Food Specialties, Delft,
carboxypeptidasea pH 4 The Netherlands
 journal of functional foods 13 (2015) 262275
l-carnosine (226.24 Da), and a mixture of Trp (204.2 Da) and Ala
(89.09 Da). The resulting estimations of molecular weight size
range were <200 Da, 2001000 Da, and >1000 Da. Blue dextran
(2000 kDa) was used to determine the column void volume.
2.7. Antihypertensive activity in spontaneously
hypertensive rats
Spontaneously hypertensive rats (SHRs) housed in the Animal
Housing Facility at the Richardson Centre of Functional Foods
and Nutraceuticals (University of Manitoba, Winnipeg, MB) were
used to determine the in vivo antihypertensive effect of T3, T3-
AC7, T3-PR7 and T3-PX7 based on the protocol described in
Girgih, Udenigwe, Li, Adebiyi, and Aluko (2011). Experiments
were carried out following the Canadian Council on Animal Care
ethics guidelines and using a protocol approved by the Uni-
versity of Manitoba Animal Protocol Management and Review
Committee (protocol number F11-015/1/2). Twenty-four male
rats about 20 weeks old and weighing 350415 g were ran-
domly assigned into six groups (n = 4 in each group): (1) T3, (2),
(3) and (4) the three 7 h exopeptidase-treated hydrolysates
(100 mg protein/kg body weight), (5) captopril (10 mg/kg body
weight, positive control), or (6) saline (negative control). All
samples were dissolved in phosphate buffered saline (pH 7.4)
and 1 mL was administered to the rats by oral gavage. At 2, 4,
6, 8 and 24 h after the single administration, the systolic blood
pressure (SBP) of SHRs was measured in triplicate by tail-cuff
plethysmography (Mouse Rat Tail Cuff Blood Pressure System,
IITC Life Science, Woodland Hills, CA, USA). The change in SBP
for each SHR was determined as the difference between SBP
at the different time points and their baseline SBP (186 1 mm
Hg), and was expressed as the mean and standard deviation
of the averaged triplicate SBP readings from four rats per group.
2.8. Sensory evaluation
Sensory evaluation was conducted with the approval of the
Behavioural Research Ethics Board of UBC (Vancouver, BC)
(Ethics Certificate Number H13-01909). Preliminary screening
of WPI and 18 WPH samples by three researchers was used as
the basis for selecting six WPH samples for evaluation by a
trained sensory panel. Panelists were recruited from the Food
Nutrition and Health program at UBC. Individuals who have
food allergies, a family history of food allergies or severe lactose
intolerance, or who are non-eaters of dairy products, were ex-
cluded from the study. The trained sensory panel consisted of
eight females and four males between 20 and 35 years of age.
One participant was Caucasian and the remaining were Asian.
Panelists participated in two training sessions, during which
they were asked to differentiate aqueous solutions of bitter stan-
dards (5, 10 and 15 mM caffeine in ddH2O) in two sets of triangle
tests, and rank bitter reference standards consisting of various
caffeine concentrations (2.515 mM caffeine dissolved in WPH
4003) in two sets of ranking tests. The commercial product WPH
4003 was selected as the base for reference standards due to
its low intensities of taste attributes. Potential product-
specific reference standards were also prepared for the purpose
of calibrating panelists in their taste assessments. The refer-
ence standards used (and their distance on the line scale, as
agreed upon by the entire panel), were as follows: 10 mM
265
caffeine for bitterness (11.0 cm), 1% monosodium glutamate
(MSG) for umami taste (12.0 cm), 2% sucrose for sweetness
(10.5 cm), 1% sodium chloride (NaCl) for saltiness (10.0 cm), and
0.4% citric acid for sourness (7.5 cm).
Sensory evaluation was conducted over two sessions and
using the procedure based on that described in Cheung and
Li-Chan (2010). Panelists tasted samples (10 g/100 mL ddH2O)
in assigned random order under red lighting to minimize any
possible visual cues, such as differences in color, among
samples. Samples (0.3 mL) were deposited directly on the back
of the tongue for tasting using a 1 mL syringe (BD; Franklin
Lakes, NJ). Panelists were instructed to wait 5 s before swal-
lowing or discarding samples, and 2 min between samples.
Bitterness, umami taste, sweetness, saltiness, sourness, and
overall acceptability were measured on 15 cm line scales with
anchors at 1 and 14 cm labeled low and high, respectively.
The reference standards were provided to panelists and used
as necessary. Water, low-sodium club soda (45 mg Na/355 mL)
(Canada Dry Motts Inc.; Mississauga, ON), and unsalted crack-
ers were provided as palate cleansers and used at least once
between samples.
2.9. Salt content
Salt content of hydrolysates was measured using a MeterLab
CDM210 conductivity meter (Radiometer Analytical SAS, Lyon,
France). The electrode was submerged into samples (ca. 15 mL
of 10% aqueous solutions), and the conductivity reading (S/cm)
recorded when the reading was stabilized. A standard curve
(0.01251.0000% NaCl) was constructed and used to deter-
mine salt content, which was expressed as % NaCl equivalent.
2.10. Amino acid analysis
Total and free amino acid contents of hydrolysates were ana-
lyzed at the Advanced Protein Technology Centre at the Hospital
for Sick Children (Toronto, ON) by reversed phase high perfor-
mance liquid chromatography with detection at 254 nm
following pre-column derivatization of hydrolysates with phenyl
isothiocyanate. Free amino acid content was assessed without
prior hydrolysis while assessment for total amino acid content
required a 24 h hydrolysis with 6 N HCl and 1% phenol at 110 C
prior to pre-column derivatization. The content of bound amino
acids was estimated as the difference between total and free
amino acid content.
2.11. Statistical analysis
The EH and ACE-inhibitory activity of hydrolysates were ana-
lyzed by a one-factor analysis of variance (ANOVA). Significant
differences (p 0.05) among means were determined using
Tukeys method. Correlation between EH and ACE-inhibitory
activity for the WPHs (n = 7) was determined by calculating the
Pearson correlation coefficient (R) and its p-value.
The effect of captopril, T3, and the three exopeptidase-
treated hydrolysates on the SBP reduction of SHRs was analyzed
by a two-factor repeated measures ANOVA with interaction.
The main effects of treatment and time (repeated factor), and
the interaction effect (treatment time) were calculated. Tukeys
method was used to determine significant differences (p 0.05)
266 journal of functional foods 13 (2015) 262275
among the five treatments at five time intervals. Saline, the
negative control, was not included in the statistical analysis
for significant differences.
The sensory data were analyzed based on the method de-
scribed by Cheung and Li-Chan (2014). A three-factor ANOVA
using a general linear model (GLM) was performed on the
sensory scores to evaluate panel performance. Panelists, sample,
and replication were analyzed as fixed effects and all main and
interaction effects were calculated. Panelists were then screened
for their individual repeatability using a two-factor ANOVA GLM
with sample and replication as main effects. Panelists with a
significant replication effect (p 0.05) had their sensory scores
removed from the overall data set. A second three-factor ANOVA
GLM confirmed an improvement in panel performance, in that
bitterness no longer had a significant panelist sample inter-
action. Since significant interaction effects remained for umami
taste, sweetness, saltiness and sourness, these taste attri-
butes were further assessed by correlating each panelists
means with the means from the other panelists. The scores
of panelists whose sample means did not significantly corre-
late (p 0.05) with the sample means from the other panelists
were removed. This revealed that the panel lacked correla-
tion in their assessment for sweetness and sourness. Thus, these
two taste attributes were not included in subsequent analy-
ses. To eliminate significant panelist effects, the sensory scores
for bitterness (n = 22, panelists = 11, replication = 2), umami taste
(n = 14, panelists = 7, replication = 2) and saltiness (n = 14, pan-
elists = 7, replication = 2) were normalized into Z-scores using
the equation as described in Mui, Durance, and Scaman (2002)
and Reid and Durance (1992):
Z = ( x X ) SD
where Z is the transformed score, x is the original score, X is
the mean of all scores for one attribute from one panelist, and
SD is the standard deviation of X for the same panelist. Both
sensory scores and normalized Z-scores were analyzed by a
three-factor ANOVA followed by determination of significant
Table 2 Extent of hydrolysis, ACE-inhibitory activity and taste of whey protein isolate (WPI) and seven whey protein
hydrolysates.
Protease used (Sample code) Extent of hydrolysis
(meq Leu/g)a
Thermoase PC10F (T3) 1.57c
Protin SD NY10 (NY3) 0.54e
Protease P Amano 6SD (PP3) 3.74a
Validase Papain Liquid (VP3) 0.88d
Protease M Amano SD (PM3) 2.18b
Protin SD AY10 (AY3) 0.84d
Maxazyme NNP DS (MX3) 0.35e
None (WPI) 0.03f
a
Expressed as -amino groups in milliequivalents of l-Leu/g of protein (meq Leu/g). Samples not sharing common lowercase letters are sig-
nificantly different (p 0.05).
b
At final assay concentration of 0.3 mg/mL. Samples not sharing common lowercase letters are significantly different (p 0.05). N.I. means
sample was not inhibitory.
c
Samples were tasted as 10% solutions by three researchers in preliminary studies. Intensities were coded as follows: no bitterness (), de-
tectable bitterness (+), mildly bitter (++), moderately bitter (+++), and highly bitter (++++).
differences (p 0.05) using Fishers least significant differ-
ence test. Sample differences based on the sensory scores for
overall acceptance were assessed using an adjusted F-value,
which was calculated using the mean square of the panel-
ist sample interaction effect as the denominator in lieu of
the mean square error. Correlation analyses were also per-
formed between mean sensory scores and results of the
compositional analyses: saltiness and umami taste with salt
content (% NaCl equivalent) (n = 6), and umami taste with free
Glu content (n = 4).
All analyses were conducted using Microsoft Excel or
Minitab 16 (Minitab Inc., State College, PA).
3. Results and discussion
3.1. Screening of WPHs for ACE-inhibitory activity and
bitterness
The EH and ACE-inhibitory activity of WPHs produced by hy-
drolyzing WPI with one of the seven commercial endoproteases
are shown in Table 2. The WPHs had various EH ranging from
0.35 to 3.74 meq Leu/g and, with the exception of MX3, all dis-
played greater than 50% ACE-inhibitory activity at a final assay
concentration of 0.3 mg/mL. A correlation between EH and ACE-
inhibitory activity of the WPHs was not observed (R = 0.3893,
n = 7, p = 0.3880), which may be due to ACE-inhibitory activity
being influenced by factors such as enzyme specificity (Mullally,
Meisel, & FitzGerald, 1997).
The unhydrolyzed WPI was not bitter, but all seven WPHs
exhibited bitterness of varying intensities (Table 2). An asso-
ciation between EH and bitterness was not observed, which
may be due, in part, to differences in the types of peptides and
free amino acids released by the different commercial
endoproteases used for hydrolysate production (Newman et al.,
2014). The most bitter hydrolysate was T3, which also had the
ACE-inhibitory Taste attributes and sensationsc
activity (%)b
Bitterness Other
75a ++++ Initially sour
69b + Film-like coating
67b + Cardboardy, sour/sweet
67b +++ Odd, sour, cardboardy
59c ++ Sour/sweet
54c + Sour, slight cardboardy
27d ++ Mild taste
N.I. Metallic, flat
 journal of functional foods 13 (2015) 262275 267

Table 3 Extent of hydrolysis, ACE-inhibitory activity, and taste of an antihypertensive whey protein hydrolysate (T3),
exopeptidase-treated T3, and four commercial products.
Samplea Extent of hydrolysis ACE inhibitory Taste attributes and sensationse
(meq Leu/g)b,c activity (%)c,d
Bitterness Other
T3 1.12f 76a ++++ Mild taste, milky,
cardboardy
Carboxypeptidase-treated
T3-AC1 1.99e 69b +++ Mild
T3-AC7 2.46cd 67b + Mild, good flavor, complex
Aminopeptidase-treated
T3-PR1 2.25d 65bc +++ Odd, cardboardy
T3-PR7 3.92b 60cd ++ Salty, umami, acceptable
T3-PX1 2.66c 70b +++ Milky, cardboardy
T3-PX7 4.55a 48e ++ Sour, good/not good
Commercial products
WPH 4003 0.01i N.I. Very mild, sweet, milky
Hilmar 8350 0.71g 16f ++++ Cracker-like, milky
Hilmar 8390 2.38d 57d +++++ Like rotten cheese
NZMP WPH 917 0.33h 21f + Thick, milky
a
A whey protein hydrolysate produced after hydrolysis with Thermoase PC10F for 3 h (T3) was treated with Accelerzyme CPG (T3-AC), Pep-
tidase R (T3-PR), or ProteAX (T3-PX). Sample codes were followed by a 1 or 7 to denote 1 or 7 h of exopeptidase treatment, respectively.
b
Expressed as -amino groups in milliequivalents of l-Leu/g of hydrolysate (meq Leu/g).
c
Samples not sharing common lowercase letters are significantly different (p 0.05).
d
At final assay concentration of 0.3 mg/mL. Samples with no inhibition are denoted as N.I.
e
Samples were tasted as 10% solutions by three researchers in preliminary studies. Intensities were coded as follows: no bitterness (), de-
tectable bitterness (+), mildly bitter (++), moderately bitter (+++), highly bitter (++++), and extremely bitter (+++++).

highest ACE-inhibitory activity and was therefore selected for
exopeptidase treatments. NY3, PP3, and VP3 also exhibited high
ACE-inhibitory activity, but had lower bitterness than T3.
However, these three WPHs exhibited other undesirable taste
and texture attributes: NY3 elicited an undesirable film-like sen-
sation while PP3 and VP3 had strong cardboardy and sour
sensations. The cardboardy sensation of WPHs has been at-
tributed to lipid oxidation products while sour taste has been
attributed to peptides (Leksrisompong et al., 2010).
3.2. ACE-inhibitory activity and preliminary taste
assessment of exopeptidase-treated T3
All exopeptidase-treated hydrolysates displayed ACE-inhibitory
activity, albeit their activities were generally lower than T3
(Table 3). EH significantly increased when the hydrolysis time
of exopeptidase treatments increased from 1 to 7 h, but only
resulted in significantly lower ACE-inhibitory activity in the case
of ProteAX-treated hydrolysates. This significant reduction in
ACE-inhibitory activity may be due to ProteAX having protein-
ase activity in addition to exopeptidase activity, as opposed to
Accelerzyme CPG and Peptidase R, which only exhibit car-
boxypeptidase and aminopeptidase activity, respectively. The
high EH of ProteAX-treated hydrolysates suggested that the pro-
teinase was active under the hydrolysis conditions used and
may have fragmented active peptides to result in lower ACE-
inhibitory activity.
In preliminary tastings, the debittering effect of
exopeptidase treatments was found to increase as hydrolysis
times were prolonged from 1 to 7 h (Table 3). In addition to
bitterness, the hydrolysates resulting from exopeptidase
treatments of 1 h had mild, milky, and cardboardy taste
sensations, similar to T3. In contrast, exopeptidase treatments
of 7 h resulted in larger decreases of bitterness as well as
increases of taste attributes such as saltiness and umami taste.
T3-AC7 and T3-PR7 were also described as good and acceptable,
respectively. These results suggested that the debittering effects
of exopeptidase treatment may be detected before the
degradation of WPH ACE-inhibitory activity. Furthermore, when
comparing aminopeptidase- and carboxypeptidase-treated
hydrolysates with similar EH (i.e., T3-AC7 and T3-PR1),
differences in ACE-inhibitory activity were not observed but
T3-AC7 had much lower bitterness. These preliminary
observations may suggest that a larger debittering effect is
achieved using carboxypeptidases instead of aminopeptidases
even when the amount of amino acids released is similar. A
possible explanation for this may be that C-terminal amino acid
residues, particularly those with hydrophobic side chains, have
more importance than N-terminal amino acid residues at
eliciting bitter taste, an association which has been predicted
by computer modeling (Kim & Li-Chan, 2006). However, these
speculations require further evaluation, since the release and
accumulation of taste-active amino acids may result in complex
taste profiles, and thus make bitterness difficult to assess
accurately by the untrained panel used in preliminary tastings.
Since the hydrolysates treated with exopeptidases for 7 h
had lower bitterness and similar ACE-inhibitory activity to hy-
drolysates treated with exopeptidases for 1 h, only the 7 h
exopeptidase-treated hydrolysates (i.e., T3-AC7, T3-PR7, and T3-
PX7), as well as T3, were included in subsequent analyses.
3.3. Size fractions of WPH and their ACE-inhibitory
activity
Size exclusion chromatography was used to fractionate T3,
which had an IC50 value of 0.15 mg/mL for ACE-inhibitory
268 journal of functional foods 13 (2015) 262275
activity, and the 7 h exopeptidase-treated samples (IC50 = 0.24,
0.27, and 0.34 for T3-AC7, T3-PR7, and T3-PX7, respectively) to
further explore the changes in peptide size distribution after
exopeptidase treatment. As shown in Fig. 1a, T3 was found
to consist predominantly of peptides with molecular
weights >1000 Da (Fraction 1), with a smaller proportion of pep-
tides between 200 and 1000 Da (Fractions 2 and 3) and virtually
no free amino acids. Exopeptidase treatments increased the
content of peptides and free amino acids in lower molecular
weight fractions (approximately <200 Da), as evident in Frac-
tions 46 of T3-AC7 (Fig. 1b), and Fractions 5 and 6 of T3-PR7
(Fig. 1c) and T3-PX7 (Fig. 1d).
All five fractions collected from T3 had ACE-inhibitory ac-
tivity, with Fractions 13 having activities of 60% at a final
assay concentration of 0.1 mg/mL (Fig. 1a). In comparison, only
one fraction in T3-AC7 (Fig. 1b) and T3-PX7 (Fig. 1d) and two
fractions in T3-PR7 (Fig. 1c) exhibited activities similar to or
higher than 60%. These fractions corresponded to peptides with
approximate molecular weights between 200 and 1000 Da in
T3-AC7 and T3-PX7 (Fraction 2), and peptides around 1000 Da
(Fraction 2) and <200 Da (Fraction 4) in T3-PR7. This molecu-
lar weight size range (i.e., <1000 Da) has been reported to exhibit
the highest ACE-inhibitory activity in other protein hydroly-
sates (Estvez et al., 2012; OLoughlin et al., 2014;
Pihlanto-Leppl et al., 2000).
On the contrary, peptides eluting after the 2001000 Da frac-
tions, which are presumably di- and possibly tri-peptides, and
collected in Fractions 4 and 5 of T3-AC7 and T3-PX7, as well
as Fraction 5 of T3-PR7, had relatively low ACE-inhibitory ac-
tivity. This may be explained by the substrate specificity of the
exopeptidases for hydrophobic amino acids (Table 1), which are
associated with ACE-inhibitory activity when present at either
terminal of di- and tri-peptides (Wu et al., 2006b). For example,
Peptidase R and ProteAX, both of which have specificity for hy-
drophobic amino acids (e.g., Ala, Leu, Phe, and Met), would
presumably degrade di-, tri- and oligo-peptides containing hy-
drophobic amino acids at the N-terminal into shorter peptides
and free amino acids. Therefore some, if not most, of the re-
maining di- and tri-peptides present in Fraction 5 of T3-PR7
and Fractions 45 in T3-PX7 may not contain hydrophobic amino
acids at the N-terminal, and are thus less likely to exhibit ACE-
inhibitory activity. Likewise, the high specificity of Accelerzyme
CPG for five hydrophobic amino acids may partly explain the
low activities observed in Fractions 4 and 5 of T3-AC7.
3.4. Antihypertensive activity of WPHs in spontaneously
hypertensive rats
A summary of the repeated measures ANOVA for the antihy-
pertensive effect of WPHs in spontaneously hypertensive rats
is shown in Table 4a. Treatment (i.e., whether the SHRs were
administered T3, T3-AC7, T3-PR7, T3-PX7, or captopril) was a
significant effect, suggesting that the five samples varied in their
antihypertensive effect. Time was also a highly significant effect,
with the magnitude of SBP reduction decreasing over time. This
was expected since ACE inhibitors and peptides are elimi-
nated from the body and therefore have transient
antihypertensive activity. Furthermore, the significant treat-
ment time interaction suggested that the rate at which SBP
returned to baseline varied among the five samples. For
example, the reduction of SBP by captopril remained similar
from 2 to 8 h after a single administration (Table 4b). In con-
trast, the antihypertensive effect of T3 was greater at 2 and 4 h
than at 6 and 8 h. Therefore, the antihypertensive activity of
captopril was more sustained than T3 during the first 8 h after
administration, although SHRs receiving T3 had a signifi-
cantly larger reduction in SBP after 2 h than those receiving
captopril. All three exopeptidase-treated hydrolysates dis-
played similar antihypertensive effects, with T3-AC7, T3-PR7,
and T3-PX7 having maximum effects of 31 2, 31 1, and
30 2 mm Hg, respectively. These hydrolysates were as ef-
fective as T3 at decreasing SBP and comparable to captopril
during the first 8 h after administration.
The three 7 h exopeptidase-treated hydrolysates and T3 had
higher antihypertensive activity than those reported for protein
hydrolysates of sweet potato (Ishiguro et al., 2012), hemp seed
(Girgih et al., 2011) and pea (Li et al., 2011) at equal or higher
doses. The WPHs in this study lowered SBP at a magnitude
similar to single oral administration of an Alcalase-hydrolyzed
WPH at 801200 mg/kg body weight (Wang et al., 2012), but were
not as effective as the whey protein and WPHs reported by
Abubakar et al. (1998), which reduced SBP when adminis-
tered at 8 mg/kg by gastric intubation. However, the
experimental hydrolysates in the current study were active
within 2 h of administration and still displayed antihyperten-
sive effect 24 h after administration, while the aforementioned
hydrolysates had greatest effect on SBP after 6 h of adminis-
tration and did not have any effect after 24 h (Abubakar et al.,
1998; Wang et al., 2012). Compared to tri-peptides sourced from
-casein, which were more potent but generally had minimal
activity 5 h after administration (Jing et al., 2014), the WPHs
in the current study had longer-lasting effects. The effect of
prolonged WPH administration on SBP reduction was not de-
termined in this study, although long-term (ca. 20 days) daily
administration of WPHs to SHRs has been reported in other
studies to decrease heart rate, plasma ACE activity, serum an-
giotensin II and plasma aldosterone levels (Fernndez-Musoles,
Manzanares, Burguete, Alborch, & Salom, 2013; Wang et al.,
2012).
3.5. Sensory evaluation of WPHs
The taste and overall acceptability of the 7 h exopeptidase-
treated hydrolysates and T3 were assessed by panelists using
descriptive sensory analysis. As well, of the four commercial
WPH products that were initially assessed during prelimi-
nary sensory evaluations for benchmarking the bitterness and
ACE-inhibitory activity of experimental hydrolysates (Table 3),
two were included in the sensory panel evaluation. As food in-
gredients, all four commercial hydrolysates act as nutrient (i.e.,
amino acid) sources and are marketed by the ingredient sup-
pliers as having minimal off-flavors and bitterness (WPH 4003),
bland taste (NZMP WPH 917 and Hilmar 8350), or flavor
enhancement benefits (Hilmar 8390). Hilmar 8390, the only
product that exhibited ACE-inhibitory activity at a level com-
parable to the experimental hydrolysates, had higher bitterness
than T3 and was therefore excluded from sensory evalua-
tion. NZMP WPH 917 was also excluded due to its distinctively
thicker consistency.
 journal of functional foods 13 (2015) 262275 269

Fig. 1 Elution profiles at 215 or 280 nm and collected fractions of T3 (a), T3-AC7 (b), T3-PR7 (c), and T3-PX7 (d) obtained
using size exclusion chromatography, and the ACE-inhibitory activity of the corresponding fractions. Fractions collected
from each hydrolysate are shown as shaded bars on elution profiles and numbered in bold. The arrows above elution
profiles denote the elution volume of four molecular weight standards: antifreeze protein (AFP, 3240 Da), vitamin B12 (VB,
1355.47 Da), l - carnosine (LC, 226.24 Da), and a mixture of tryptophan (Trp, 204.2 Da) and alanine (Ala, 89.09 Da). The ACE-
inhibitory activities (%) of fractions are shown in the bar graph below each profile.
270 journal of functional foods 13 (2015) 262275

Table 4a Summary of mean squares, F-values and Table 5 Summary of p-values, F-values and mean
p-values of a two-factor ANOVA GLM with repeated square error terms for panelist, sample, replication, and
measures on one factor (time), on the antihypertensive their interactions from a three-factor ANOVA GLM.
activity of captopril, T3 and 7 h exopeptidase-treated T3. Source p-valuea
a
Factor Mean squares F-value p-value of
Bitternessb Umami Saltinessb Overall
variation
Treatment 54.71 4.96 0.010 tasteb acceptability
Time 832.01 125.92 0.000
Panelist (P) 1.000 1.000 1.000 0.255
Treatment Time 27.84 4.21 0.000
Sample (S) 0.000 0.000 0.000 0.000
a
Significant p-values (p 0.05) are shown in bold font. Replication (R) 0.772 0.469 0.115 0.912
PS 0.067 0.485 0.838 0.000
PR 0.963 0.587 0.512 0.806
SR 0.958 0.297 0.059 0.900
Z-score transformation of sensory scores prior to statisti-
cal analysis was applied for standardization of the data (i.e., Sample 41.20 50.85 36.96 18.82
F-value (6.39)c
representing all sensory scores on the same scale), thereby
Mean square 0.3588 0.2360 0.3011 5.416
eliminating significant panelist effects (Mui et al., 2002; Reid
error
& Durance, 1992). Although panelist effects are common in
a
Six whey protein hydrolysate samples were assessed by 11, 7, 7
sensory research and generally attributed to physiological dif-
and 12 panelists for bitterness, umami taste, saltiness and overall
ferences among panelists as well as differences and
acceptability, respectively, as explained in the text. Significant effects
inconsistencies in line scale usage (Cliff, Stanich, Edwards, & (p 0.05) are shown in bold font.
Saucier, 2012; Moon & Li-Chan, 2007; Song et al., 2010), a sig- b
ANOVA performed on Z-scores.
nificant panelist effect should ideally not occur in analytical c
An adjusted F-value of 6.39 was calculated using the mean square
sensory evaluation (OMahony, 1997). The results of a three- of the panelist sample interaction effect (15.952) as the denomi-
factor ANOVA GLM on the Z-transformed scores (bitterness, nator. The sample effect was significant (p 0.001).

umami taste, saltiness) and overall acceptability scores are

shown in Table 5. Significant sample effects were found for all
three taste attributes, suggesting that the hydrolysates had dis-
tinguishable intensities of bitterness, umami taste and saltiness.
It should be noted, however, that since Z-scores were used for
statistical analysis, the sample differences observed are only
relative to one another, and indicated by their positive or nega-
tive deviation around a mean of zero (Table 6). Sensory scores,
on the other hand, provide information on the taste inten-
sity of a sample (e.g., on a scale from 0 to 15) and its relative
distance from a known reference standard. Interestingly, the
intensity of the bitter standard used in the current study (10
mM or 0.19% caffeine in WPH 4003 placed at 11 cm on a 15 cm
line scale) was found to be similar to one used by Liu, Jiang
and Peterson (2014) (0.15% caffeine in water placed at 10 on a
15 point scale), suggesting that the panelists between the two
studies had similar perceptions of bitterness.
The trained panel found T3 to have the highest bitterness
of all hydrolysates tasted, with bitterness rated slightly less than
the reference standard (10 mM caffeine in WPH 4003), and was
significantly more bitter than the exopeptidase-treated hydro-
lysates (Table 6). Of the exopeptidase-treated hydrolysates, both
T3-AC7 and T3-PR7 were significantly less bitter than T3-PX7.
This may be due to the proteinase activity in ProteAX gener-
ating high contents of low molecular weight peptides, which
are correlated to bitterness in hydrolysates (Leksrisompong
et al., 2010). The two hydrolysates treated with only exopep-
tidases (i.e., T3-AC7 and T3-PR7) had bitterness similar to
Hilmar 8350, a commercial product described by its sup-
plier as having bland taste. This suggested that exopeptidase
treatment with either an aminopeptidase or carboxypepti-
dase can potentially decrease the bitterness of ACE-inhibiting
hydrolysates to levels comparable to commercial WPHs.
Exopeptidase treatment also influenced the other taste at-
tributes of the exopeptidase-treated hydrolysates. Umami taste,

Table 4b Mean systolic blood pressure reduction in spontaneously hypertensive rats over 24 h after a single
administration of saline, captopril, T3, T3-AC7, T3-PR7 or T3-PX7.
Sampleb Mean change in systolic blood pressure (mm Hg)a at various time intervals after a
single administration
2h 4h 6h 8h 24 h
Saline 5 2 3 3 2
Captopril 27 defgh 28 efghi 24 cdefg 23 cde 20 bcd
T3 35 i 34 hi 25 cdefg 23 cde 16 ab
T3-AC7 29 efghi 31 ghi 28 efgh 24 cdef 12 a
T3-PR7 27 defg 31 ghi 27 efgh 19 abc 12 a
T3-PX7 30 fghi 27 defgh 24 cdef 18 abc 12 a
a
Means (n = 4) not sharing common lowercase letters are significantly different (p 0.05).
b
A whey protein hydrolysate produced after hydrolysis with Thermoase PC10F for 3 h (T3) was treated for 7 h with Accelerzyme CPG (T3-
AC7), Peptidase R (T3-PR7) or ProteAX (T3-PX7) and administered to spontaneously hypertensive rats at 100 mg/kg body weight by oral gavage.
Captopril was administered at 10 mg/kg body weight.
 journal of functional foods 13 (2015) 262275 271

Table 6 Mean sensory and Z-scores for bitterness, umami taste, saltiness and overall acceptability of T3, exopeptidase-
treated T3, and two commercial products Hilmar 8350 and WPH 4003.
Samplea Taste attributeb Overall acceptabilityb
Bitterness Umami taste Saltiness
c d c d
Sensory Z-scores Sensory Z-scores Sensoryc Z-scoresd
T3 10.4 1.1a 4.2 0.5b 3.5 0.3c 3.8d
T3-AC7 6.5 0.1c 9.9 1.0a 7.4 1.2a 6.7bc
T3-PR7 7.1 0.1c 9.0 0.8a 5.0 0.3b 6.4bc
T3-PX7 8.2 0.5b 8.5 0.7a 6.0 0.6b 5.7cd
Hilmar 8350 5.8 0.2c 2.8 0.9bc 2.2 0.8d 8.3ab
WPH 4003 2.3 1.4d 1.9 1.1c 1.4 1.1d 9.7a
a
Experimental samples consisting of a whey protein hydrolysate produced after hydrolysis with Thermoase PC10F for 3 h (T3) and treatment
with Accelerzyme CPG (T3-AC7), Peptidase R (T3-PR7) or ProteAX (T3-PX7) for 7 h were tasted as 10% solutions in duplicate. Two commer-
cial products Hilmar 8350 and WPH 4003 were also tasted.
b
Means in the same column not sharing common lowercase letters are significantly different (p 0.05) based on the analysis of bitterness,
umami taste, saltiness and overall acceptability by 11, 7, 7 and 12 panelists, respectively.
c
Significant sample differences were not calculated due to significant panelist and interaction effects when the data were analyzed by a three-
factor ANOVA using sample, panelist, and replication as main effects.
d
Z-scores that are above and below the mean have positive and negative values, respectively.

while minimally present in T3 and the two commercial prod-
ucts, was significantly higher in exopeptidase-treated
hydrolysates (Table 6). Umami taste describes a savory, satis-
fying taste and is often associated with the taste improvement
ability of monosodium glutamate (MSG) (Beauchamp, 2009).
Therefore, exopeptidase treatment appeared to enhance the
savory taste of T3 in addition to having a debittering effect. It
is possible that the increase in umami taste contributed to the
debittering effect of exopeptidase treatment, since the addi-
tion of MSG (100 mM) has been reported to decrease the
bitterness of a 10% WPH (Leksrisompong et al., 2012). The
exopeptidase-treated hydrolysates also had significantly higher
saltiness than T3, with T3-AC7 having the highest saltiness of
the samples assessed. The saltiness of the WPHs was found
to be significantly correlated to their salt contents (R = 0.9839,
n = 6, p = 0.0004). Of the six WPHs, T3-AC7 had the highest NaCl
equivalent of 0.56%. T3-PX7, T3-PR7, T3, Hilmar 8350 and WPH
4003 had NaCl equivalents of 0.39, 0.35, 0.27, 0.20, and 0.10%,
respectively.
While panelists chose to include sweetness and sourness
as descriptors for the hydrolysates, the assessment of the
sensory data for panel consistency revealed that there was a
lack of agreement among panelists in their evaluation of these
two attributes. This may be due to the inherently low sweet-
ness and sourness of the experimental and commercial
hydrolysates, which had mean (n = 144) sensory scores of
3.4 2.2 and 2.1 1.3, respectively. Therefore, these two taste
attributes may not be relevant descriptors of the experimen-
tal and commercial hydrolysates evaluated in the current study.
3.6. Correlations of taste attributes and overall
acceptability
In addition to taste attributes, the overall acceptability of the
experimental and commercial hydrolysates was assessed. As
shown in Table 5, the F-value for the sample effect was recal-
culated using the mean square of the interaction effect (Goniak
& Noble, 1987; King, Dunn, & Heymann, 2013) (MSps = 15.952,
Fadj = 6.39, p 0.001) to overcome the significant panel-
ist sample interaction. T3-AC7 and T3-PR7 had significantly
higher overall acceptability than T3 and were comparable to
Hilmar 8350 (Table 6). The increased overall acceptability of
the exopeptidase-treated hydrolysates may be due to their
changes in individual taste attributes. For example, exopep-
tidase treatment of T3 increased the intensity of umami taste
and decreased bitterness; these taste attributes are associ-
ated with food acceptance and rejection, respectively
(Beauchamp, 2009; Temussi, 2011). The lack of significant dif-
ferences in overall acceptability among exopeptidase-treated
hydrolysates may be attributed to the small panel size (n = 12).
While this size serves to estimate the general direction of
overall acceptability, it is not representative of the general
consumer population and may not be sufficient to establish
minor sample differences.
Bitterness was the only taste attribute significantly corre-
lated to overall acceptability, to which it correlated negatively
(R = 0.9761, n = 6, p = 0.0008) (Table 7). This demonstrated that
the low overall acceptability of T3 was largely due to its bitter
taste and suggested that overall acceptability can be in-
creased by decreasing the bitterness. Since consumer
acceptance is required for functional food products to be suc-
cessful (Barrios et al., 2008; Verbeke, 2006), debittering of bitter
hydrolysates prior to incorporation into functional foods would
be highly beneficial.
Umami taste was positively correlated to saltiness (R = 0.9604,
n = 6, p = 0.0023) (Table 7). Panelists may be prone to associ-
ate umami and salty tastes due to the likelihood of tasting both
attributes simultaneously during meals (Ikeda, 2002). In the
current study, the hydrolysate salt contents (i.e., % NaCl equiva-
lent) were significantly correlated with umami taste (R = 0.9230,
n = 6, p = 0.0087). This may suggest that the salt present in hy-
drolysates interacted with the free glutamate produced after
exopeptidase treatment, which led to an enhancement of
umami taste and saltiness that was difficult to differentiate.
Of the experimental hydrolysates, T3-AC7 had the highest salt
content, umami taste, and saltiness.
272 journal of functional foods 13 (2015) 262275

Table 7 Correlation matrix displaying the Pearson correlation coefficient and p-value among three taste attributes and
overall acceptability.
Attribute Pearson correlation coefficient (p-value)a
Bitterness Umami taste Saltiness Overall acceptability
Bitterness 1.0000 (0.0000) 0.3963 (0.4367) 0.4633 (0.3548) 0.9761 (0.0008)
Umami taste 1.0000 (0.0000) 0.9604 (0.0023) 0.4245 (0.4015)
Saltiness 1.0000 (0.0000) 0.4919 (0.3216)
Overall acceptability 1.0000 (0.0000)
a
Correlations were conducted using mean sensory scores. Significant correlations (p 0.05) are shown in bold font (n = 6, df = 4).

3.7. Amino acid analysis
The total and free amino acid contents of T3 and the 7 h
exopeptidase-treated hydrolysates are shown in Fig. 2. The origi-
nal hydrolysate, T3, had high contents of Glu, Leu, Asp, and Lys,
the four of which accounted for 50% of the total amino acids
present. These values were similar to those of the starting ma-
terial used for hydrolysate production (NZMP WPI 895), as
well as to WPHs reported elsewhere (Sinha, Radha, Prakash,
& Kaul, 2007). As suggested by its elution profile at 214 nm
(Fig. 1a), T3 contained very low amounts of free amino acids
(0.561 g/100 g sample), consisting mostly of Trp and Val.
The aminopeptidase-treated sample T3-PX7, and to a lesser
extent T3-PR7, released the majority of Leu, Ile, Val, Phe and
Arg as free amino acids from their peptide-bound form (Fig. 2),
as may be expected based on their enzyme specificities (Table 1).
As previously stated, terminal hydrophobic amino acids have
been correlated with ACE-inhibitory activity. In addition, hy-
drophobic amino acids such as the branched-chain Ile, Leu and
Val and the aromatic ring containing Phe, Trp, and Tyr, as well

Fig. 2 Bound and free contents of nonpolar (a) and polar amino acids (b) in the hydrolysate produced by Thermoase PC10F
(T3), and T3 treated with Accelerzyme CPG (T3-AC7), Peptidase R (T3-PR7), or ProteAX (T3-PX7). Stacked bars depict bound
(top bars) and free (bottom bars) amino acid contents, the sum of which account for the total amino acid content. Four
samples were assessed (from left to right): T3 (solid pattern), T3-AC7 (diagonal striped pattern), T3-PR7 (vertical striped
pattern), and T3- PX7 (checkered pattern). Asx represents the sum of Asp and Asn, while Glx represents the sum of Glu and
Gln. Their respective free amino acid contents (g/100 g sample) in T3, T3-AC7, T3- PR7, and T3-PX7 were as follows: 0.02,
1.25, 0.90, and 1.00 for Asp; 0.00, 0.15, 0.59, and 1.07 for Asn; 0.00, 2.57, 1.88, and 1.35 for Glu; and 0.00, 0.87, 1.30, and 0.78
for Gln. Cys and Trp contents were underestimated due to their sensitivity to acid hydrolysis.
 journal of functional foods 13 (2015) 262275
as the basic amino acid Arg are bitter in their natural
l-configuration (Bassoli, Borgonovo, Caremoli, & Mancuso, 2014;
Li-Chan & Cheung, 2010). While increased contents of these
amino acids may have been expected to impart bitterness in
hydrolysates, the exopeptidase-treated hydrolysates were sig-
nificantly less bitter than T3 (Table 6). This may be due to
hydrophobic amino acids having higher bitterness when present
in peptide-bound form than as free amino acids (Matoba & Hata,
1972). Therefore, hydrolysates can be debittered despite the
release of free, bitter-tasting amino acids after exopeptidase
treatment. Hydrophobic C-terminal amino acids were pre-
dicted to correlate with bitterness in peptides (Kim & Li-Chan,
2006) and thus their removal may be expected to result in larger
decreases of bitterness than the removal of N-terminal hy-
drophobic amino acids. Although significant differences in
bitterness were not observed between T3-AC7 and T3-PR7, they
were both significantly less bitter than T3-PX7. However, it is
difficult to attribute these differences in bitterness solely on
the removal of N- versus C-terminal amino acids since other
confounding factors, such as dissimilar extent of hydrolysis and
the release of different amino acids, may modulate the bitter
perception.
While the exopeptidase-catalyzed release of bitter amino
acids may have contributed to an overall debittering effect, the
release of free Glu, which exists in its umami-tasting salt form
glutamate at neutral pH, may be associated with the increased
umami taste observed in exopeptidase-treated hydrolysates.
High contents of Glu were postulated to contribute more than
peptides to the umami taste of protein hydrolysates (Temussi,
2011). Indeed, the free Glu content was significantly correlated
to umami taste with R of 0.9695 (n = 4; p = 0.0305).
4. Conclusion
Exopeptidase treatment of a Thermoase PC10F-produced whey
protein hydrolysate resulted in hydrolysates that were signifi-
cantly lower in bitterness and still exhibited ACE-inhibitory
activity in vitro. The antihypertensive effects of the experi-
mental hydrolysates were confirmed in spontaneously
hypertensive rats, in which the three 7 h exopeptidase-
treated hydrolysates (100 mg/kg body weight) were as effective
as captopril (10 mg/kg body weight) at reducing systolic blood
pressure for the first 8 h after administration, and had effects
lasting 24 h. Fractionation of the debittered hydrolysates by size
exclusion chromatography revealed that the ACE-inhibitory ac-
tivity of exopeptidase-treated hydrolysates consisted
predominantly of peptides with molecular weights in the range
of approximately 2001000 Da. In contrast, the <200 Da frac-
tions, presumably consisting of di- and tri-peptides, had
relatively lower ACE-inhibitory activity that may be ex-
plained by the specificity of exopeptidases for hydrophobic
amino acids. The release of terminal hydrophobic amino acids
may also contribute to the significant decreases in bitterness
after treatment with Accelerzyme CPG or Peptidase R, while
increasing free Glu contents may explain the increased umami
tastes observed in exopeptidase-treated hydrolysates. Since bit-
terness was the only taste attribute significantly correlated to
overall acceptability, an effective debittering method for bitter-
273
tasting whey protein hydrolysates may be essential for their
successful incorporation into functional foods. The results of
this study demonstrate that exopeptidase treatment with either
aminopeptidases or carboxypeptidases has potential use for
such purposes.
Acknowledgements
Funding for this research was provided through a Discovery
Grant (Council Grant No. 121822-11) from the Natural Sci-
ences and Engineering Research Council of Canada (NSERC).
We thank Fonterra Co-operative Group for the donation of whey
protein isolate, Caldic Canada Inc., PGP International and Hilmar
Ingredients for the donation of commercial whey protein hy-
drolysates, and Amano Enzyme U.S.A., Ltd. and DSM Food
Specialties B.V. for the donation of the proteases and exopep-
tidases. LKY Cheung also thanks the Canadian Dairy
Commission for a scholarship and travel funds.
REFERENCES
Abubakar, A., Saito, T., Kitazawa, H., Kawai, Y., & Itoh, T. (1998).
Structural analysis of new antihypertensive peptides derived
from cheese whey protein by Proteinase K digestion. Journal of
Dairy Science, 81, 31313138.
Barrios, E. X., Bayarri, S., Carbonell, I., Izquierdo, L., & Costell, E.
(2008). Consumer attitudes and opinions toward functional
foods: A focus group study. Journal of Sensory Studies, 23, 514
525.
Bassoli, A., Borgonovo, G., Caremoli, F., & Mancuso, G. (2014). The
taste of D- and L-amino acids: In vitro binding assays with
cloned human bitter (TAS2Rs) and sweet (TAS1R2/TAS1R3)
receptors. Food Chemistry, 15, 2733.
Beauchamp, G. K. (2009). Sensory and receptor responses to
umami: An overview of pioneering work. The American Journal
of Clinical Nutrition, 90, 723S737S.
Cheison, S. C., Wang, Z., & Xu, S.-Y. (2007). Use of macroporous
adsorption resin for simultaneous desalting and debittering
of whey protein hydrolysates. International Journal of Food
Science and Technology, 42, 12281239.
Cheung, I. W. Y., & Li-Chan, E. C. Y. (2010). Angiotensin-I-
converting enzyme inhibitory activity and bitterness of
enzymatically-produced hydrolysates of shrimp (Pandalopsis
dispar) processing byproducts investigated by Taguchi design.
Food Chemistry, 122, 10031012.
Cheung, I. W. Y., & Li-Chan, E. C. Y. (2014). Application of taste
sensing system for characterization of enzymatic
hydrolysates from shrimp processing by-products. Food
Chemistry, 145, 10761085.
Cliff, M. A., Stanich, K., Edwards, J. E., & Saucier, C. T. (2012).
Adding grape seed extract to wine affects astringency and
other sensory attributes. Journal of Food Quality, 35, 263271.
Estvez, N., Fucios, P., Sobrosa, A. C., Pastrana, L., Prez, N., &
Ra, M. L. (2012). Modeling the angiotensin-converting
enzyme inhibitory activity of peptide mixtures obtained from
cheese whey hydrolysates using concentration-response
curves. Biotechnology Progress, 28, 11971206.
Fernndez-Musoles, R., Manzanares, P., Burguete, M. C., Alborch,
E., & Salom, J. B. (2013). In vivo angiotensin I-converting
enzyme inhibition by long-term intake of antihypertensive
lactoferrin hydrolysate in spontaneously hypertensive rats.
Food Research International, 54, 627632.
274 journal of functional foods 13 (2015) 262275
Fu, J., Li, L., & Yang, X.-Q. (2011). Specificity of carboxypeptidases
from Actinomucor elegans and their debittering effect on
soybean protein hydrolysates. Applied Biochemistry and Bio-
technology, 165, 12011210.
Ge, S.-J., & Zhang, L.-X. (1996). The immobilized porcine pancre-
atic exopeptidases and its application in casein hydrolysates
debittering. Applied Biochemistry and Biotechnology, 59, 159165.
Girgih, A. T., Udenigwe, C. C., Li, H., Adebiyi, A. P., & Aluko, R. E.
(2011). Kinetics of enzyme inhibition and antihypertensive
effects of hemp seed (Cannabis sativa L.) protein hydrolysates.
Journal of the American Oil Chemists Society, 88, 17671774.
Goniak, O. J., & Noble, A. C. (1987). Sensory study of selected
volatile sulfur compounds in white wine. American Journal of
Enology and Viticulture, 38, 223227.
Hernndez-Ledesma, B., Contreras, M. M., & Recio, I. (2011). An-
tihypertensive peptides: Production, bioavailability and incor-
poration into foods. Advances in Colloid and Interface Science,
165, 2335.
Ikeda, K. (2002). New seasonings. Chemical Senses, 27, 847849.
Ishiguro, K., Sameshima, Y., Kume, T., Ikeda, K., Matsumoto, J., &
Yoshimoto, M. (2012). Hypotensive effect of a sweetpotato
protein digest in spontaneously hypertensive rats and
purification of angiotensin I-converting enzyme inhibitory
peptides. Food Chemistry, 131, 774779.
Jing, P., Qian, B., He, Y., Zhao, X., Zhang, J., Zhao, D., Lv, Y., & Deng,
Y. (2014). Screening milk-derived antihypertensive peptides
using quantitative structure activity relationship (QSAR)
modelling and in vitro/in vivo studies on their bioactivity.
International Dairy Journal, 35, 95101.
Kim, H.-O., & Li-Chan, E. C. Y. (2006). Quantitative structure-
activity relationship study of bitter peptides. Journal of
Agricultural and Food Chemistry, 54, 1010210111.
King, E. S., Dunn, R. L., & Heymann, H. (2013). The influence of
alcohol on the sensory perception of red wines. Food Quality
and Preference, 28, 235243.
Leksrisompong, P., Gerard, P., Lopetcharat, K., & Drake, M. (2012).
Bitter taste inhibiting agents for whey protein hydrolysate
and whey protein hydrolysate beverages. Journal of Food
Science, 77, S282S287.
Leksrisompong, P. P., Miracle, R. E., & Drake, M. (2010).
Characterization of flavor of whey protein hydrolysates.
Journal of Agricultural and Food Chemistry, 58, 63186327.
Li, H., Prairie, N., Udenigwe, C. C., Adebiyi, A. P., Tappia, P. S.,
Aukema, H. M., Jones, P. J. H., & Aluko, R. E. (2011). Blood
pressure lowering effect of a pea protein hydrolysate in
hypertensive rats and humans. Journal of Agricultural and Food
Chemistry, 59, 98549860.
Li-Chan, E. C. Y., & Cheung, I. W. Y. (2010). Flavor-active properties
of amino acids, peptides, and proteins. In Y. Mine, E. Li-Chan,
& B. Jiang (Eds.), Bioactive proteins and peptides as functional
foods and nutraceuticals (pp. 341358). Iowa: Blackwell
Publishing and Institute of Food Technologists.
Liceaga-Gesualdo, A. M., & Li-Chan, E. C. Y. (1999).
Functional properties of fish protein hydrolysate from
herring (Clupea harengus). Journal of Food Science, 64,
10001004.
Liu, X., Jiang, D., & Peterson, D. G. (2014). Identification of bitter
peptides in whey protein hydrolysate. Journal of Agricultural
and Food Chemistry, 62, 57195725.
Matoba, T., & Hata, T. (1972). Relationship between bitterness of
peptides and their chemical structures. Agricultural and
Biological Chemistry, 36, 14231431.
Moon, S.-Y., & Li-Chan, E. C. Y. (2007). Changes in aroma
characteristics of simulated beef flavour by soy protein isolate
assessed by descriptive sensory analysis and gas
chromatography. Food Research International, 40, 12391248.
Mui, W. W. Y., Durance, T. D., & Scaman, C. H. (2002). Flavor and
texture of banana chips dried by combinations of hot air,
vacuum, and microwave processing. Journal of Agricultural and
Food Chemistry, 50, 18831889.
Mullally, M. M., Meisel, H., & FitzGerald, R. J. (1997). Angiotensin-I-
converting enzyme inhibitory activities of gastric and
pancreatic proteinase digests of whey proteins. International
Dairy Journal, 7, 299303.
Newman, J., Egan, T., Harbourne, N., ORiordan, D., Jacquier, J. C.,
& OSullivan, M. (2014). Correlation of sensory bitterness in
dairy protein hydrolysates: Comparison of prediction models
built using sensory, chromatographic and electronic tongue
data. Talanta, 126, 4653.
Nishiwaki, T., Yoshimizu, S., Furuta, M., & Hayashi, K. (2002).
Debittering of enzymatic hydrolysates using an
aminopeptidase from the edible basidiomycete Grifola
frondosa. Journal of Bioscience and Bioengineering, 93,
6063.
OLoughlin, I. B., Murray, B. A., FitzGerald, R. J., Brodkorb, A., &
Kelly, P. M. (2014). Pilot-scale production of hydrolysates with
altered bio-functionalities based on thermally-denatured
whey protein isolate. International Dairy Journal, 34, 146152.
OMahony, M. (1997). Comments on Ns, Lansgrud & Steinsholt.
Food Quality and Preference, 9, 165.
Pihlanto-Leppl, A., Koskinen, P., Piilola, K., Tupasela, T., &
Korhonen, H. (2000). Angiotensin I-converting enzyme
inhibitory properties of whey protein digests: Concentration
and characterization of active peptides. Journal of Dairy
Research, 67, 5364.
Pripp, A. H., & Ard, Y. (2007). Modelling relationship between
angiotensin-(I)-converting enzyme inhibition and the bitter
taste of peptides. Food Chemistry, 102, 880888.
Public Health Agency of Canada (2009). Tracking heart disease and
stroke in Canada. Ottawa: Public Health Agency of Canada.
Reid, R. A., & Durance, T. D. (1992). Textural changes of canned
chum salmon related to sexual maturity. Journal of Food
Science, 57, 13401342.
Ritter, J. M. (2011). Angiotensin converting enzyme inhibitors and
angiotensin receptor blockers in hypertension. BMJ (Clinical
Research Ed.), 342, 18.
Sagardia, I., Roa-Ureta, R. H., & Bald, C. (2013). A new QSAR
model, for angiotensin I-converting enzyme inhibitory
oligopeptides. Food Chemistry, 136, 13701376.
Saha, B. C., & Hayashi, K. (2001). Debittering of protein
hydrolyzates. Biotechnology Advances, 19, 355370.
Sinha, R., Radha, C., Prakash, J., & Kaul, P. (2007). Whey protein
hydrolysate: Functional properties, nutritional quality and
utilization in beverage formation. Food Chemistry, 101, 1484
1491.
Song, S., Zhang, X., Hayat, K., Huang, M., Liu, P., Karangwa, E., Gu,
F., Jia, C., Xia, S., Xiao, Z., & Niu, Y. (2010). Contribution
of beef base to aroma characteristics of beeflike process
flavour assessed by descriptive sensory analysis and gas
chromatography olfactometry and partial least squares
regression. Journal of Chromatography. A, 1217, 7788
7799.
Tan, J., Tian, F., Lv, Y., Liu, W., Zhong, L., Liu, Y., & Yang, L. (2013).
Integration of QSAR modelling and QM/MM analysis to
investigate functional food peptides with antihypertensive
activity. Molecular Simulation, 39, 10001006.
Temussi, P. A. (2011). The good taste of peptides. Journal of Peptide
Science, 18, 7382.
U.S. Food and Drug Administration. (2010). GRAS Notification No.
GRN 000345. GRAS notification for carboxypeptidase from a
genetically modified strain of Aspergillus niger. <http://www
.fda.gov/ucm/groups/fdagov-public/@fdagov-foods-gen/
documents/document/ucm269546.pdf> Accessed 15.01.07.
Verbeke, W. (2006). Functional foods: Consumer willingness to
compromise on taste for health? Food Quality and Preference,
17, 126131.
 journal of functional foods 13 (2015) 262275
Wang, L., Mao, X., Cheng, X., Xiong, X., & Ren, F. (2010). Effect of
enzyme type and hydrolysis conditions on the in vitro
angiotensin I-converting enzyme inhibitory activity and ash
content of hydrolysed whey protein isolate. International
Journal of Food Science and Technology, 45, 807812.
Wang, X., Wang, J., Lin, Y., Ding, Y., Wang, Y., Cheng, X., & Lin, Z.
(2011). QSAR study on angiotensin-converting enzyme
inhibitor oligopeptides based on a novel set of sequence
information descriptors. Journal of Molecular Modeling, 17, 1599
1606.
Wang, X., Wang, L., Cheng, X., Zhou, J., Tang, X., & Mao, X.-Y.
(2012). Hypertension-attenuating effect of whey protein
hydrolysate on spontaneously hypertensive rats. Food
Chemistry, 134, 122126.
Weber, M. A., & Messerli, F. H. (2008). Angiotensin-converting
enzyme inhibitors and angioedema: Estimating the risk.
Hypertension, 51, 14651467.
275
World Health Organization (2013). A global brief on hypertension
(Document number WHO/DCO/WHD/2013.2). Geneva: World
Health Organization.
Wu, J., Aluko, R. E., & Nakai, S. (2006a). Structural requirements of
angiotensin I-converting enzyme inhibitory peptides:
Quantitative structure-activity relationship modeling of
peptides containing 410 amino acid residues. QSAR &
Combinatorial Science, 25, 873880.
Wu, J., Aluko, R. E., & Nakai, S. (2006b). Structural requirements of
angiotensin I-converting enzyme inhibitory peptides:
Quantitative structure-activity relationship study of di- and
tripeptides. Journal of Agricultural and Food Chemistry, 54, 732
738.