Relationships of Hydrophobicity to Emulsifying
of Heat Denatured Proteins
LEANDROS P. VOUTSINAS, ELAINE
ABSTRACT
Effects of heating on the emulsifying properties of selectedfood
proteins and the protein surfacehydrophobicity (So) asa predictor
of thesepropertieswere investigated.The emulsifyingpropertiesof
the proteins studied were differently affected by heating. Heat-
denaturation was not always accompaniedby loss of emulsifying
properties, but, on the contrary, in some casesresulted in great
improvement.The emulsifying properties could welI be predicted
solely on the basisof So level but not on the basisof solubihty level,
which indicated that So is a very important property determining
protein functionality. However,the emulsifying activity, emulsion
stability and fat binding capacity of the proteins studied could be
explained and more accurately predicted using So and solubihty
together.
INTRODUCTION
TO BE USEFUL and successful in food applications, pro-
teins in addition to providing essential amino acids, should
ideally possessseveral desirable characteristics referred to as
functional properties (Wang and Kinsella, 1976). Moreover,
according to Johnson (1970) the functional and physical
properties, rather than the nutritional value, of protein in
protein-containing products will largely determine their
acceptability as ingredients in prepared foods.
The ability of protein to aid the formation and stabiliza-
tion of emulsions is critical for many applications in
chopped, comminuted meats, cake batters, coffee whiteners,
milk, mayonnaise, salad dressings, and frozen desserts. In
these products varying emulsifying and stabilizing capacities
are required because of the differing composition and
stresses to which these products are subjected (Kinsella,
1979). Moreover, the ability of proteins to bind fats is a
very important functional property for such applications
as meat replacers and extenders, principally because it
enhances flavor retention and reputedly improves mouth-
feel (Kinsella, 1976). Soy proteins have been added to
ground meats to promote fat absoption or fat binding, and
thus decrease cooking losses and maintain dimensional
stability in the cooked product (Wolf and Cowan, 1975).
To evaluate the emulsifying properties of a protein its
solubility profile is usually determined, because of its use-
fulness as an excellent index of protein functionality (Kin-
sella, 1976). Good solubility can markedly expand poten-
tial applications of a protein (Kinsella, 1976). Denaturation,
on the other hand, implicates losses of protein functional
properties and is usually measured as a loss of solubility
(Nakai and Powrie, 198 1). Generally, surfactant properties
are related to the aqueous solubility of proteins (Kinsella,
1976). A positive correlation between solubility and the
ability of a protein to emulsify and stabilize an emulsion
has been reported in many studies (Crenwelge et al., 1974;
Pearson et al., 196.5; Swift and Sulzbacher, 1963; Volkert
and Klein, 1979; Yasumatsu et al., 1972). Many authors
Authors Voutsinas, Cheung and Nakai are affiliated with the Dept.
of Food Science, The Univ. of British Columbia, Vancouver, British
Columbia, Canada V6T 2A2.
26-JOURNAL OF FOOD SCIENCE-Volume 48 (1983)
Properties
CHEUNG, and SHURYO NAKAI
point to evidence, however, that emulsifying properties and
solubility are not well correlated (Aoki et al., 1980; McWat-
ters and Cherry, 1975; McWatters and Holmes, 1979a,
1979b; Smith et al., 1973; Wang and Kinsella, 1976).
The protein hydrophobicity has been lately receiving
much attention since the hydrophobic interactions are
considered to play important roles in the functional prop-
erties of food proteins (Kato et al., 1981 ; Kinsella, 1979).
Keshavarz and Nakai (1979) reported a significant correla-
tion between surface hydrophobicity (determined by
hydrophobic chromatography and hydrophobic partition
techniques) and interfacial tension of the proteins studied.
Kato and Nakai (1980) subsequently reported that the
surface kydrophobicity (determined fluorometrically)
showed significant correlations with interfacial tension and
emulsifying activity of the proteins studied. Their results
suggest that the emulsification of oil with protein can be
explained using the concept of protein hydrophobicity.
Nakai et al. (1980b) also reported that the effective (sur-
face) hydrophobicity showed good correlations with inter-
facial tension and emulsifying activity of the plant proteins
studied. It is noteworthy that these authors observed a
closer correlation of emulsification capacity with hydro-
phobicity than with solubility.
While many factors influence the performance of pro-
teins in food systems, heat treatment is one of the most
important and is very often used during the processing of
protein products. This study was instigated by the observa-
tion that the emulsifying capacity of soy protein was not
adversely affected even by texturization which caused a loss
in solubihty. The objectives of this paper, therefore, were to
determine the effect of heating on the emulsifying proper-
ties of selected food proteins, and to assessthe value of
surface hydrophobicity as a predictor of the emulsifying
properties of these proteins.
MATERIALS&METHODS
Materials
Bovine serum albumin (#A-4503), p-lactoglobulin (#L6879
from milk) and ovalbumin (#A-5503) were ail purchasedfrom
SigmaChemicals(St. Louis, MO). Soy protein isolatewasobtained
from General Mills, Inc., (Minneapolis,MN). Promine-Dwas pur-
chasedfrom Central Soya Co. (Chicago,IL). Pea protein isolate
(a) (M412-046),Century cultivar field pea, was obtainedfrom POS
Pilot Plant Corp., Univ. of Saskatchewan(Saskatoon,Sask.).Pea
protein isolate (b), Pro-PulseWlOO, was obtained from Griffith
Laboratories Ltd. (Scarborough,Ont.). Vital wheat gluten, Whet-
pro 75%, was suppliedby Industrial Grain ProductsLtd. (Thunder
Bay, Ont.). Canola Protein isolate was preparedby the method of
Nakai et al. (198Oa).The protein content of isolatebatches(a) and
(b) was 84% and 88%, respectively.Sunflower protein isolatewas
preparedby the method of Nakai et al. (1980a).Wheyprotein con-
centrate (75%) was obtained from Sodispro Technol. (St. Hya-
&the, Que.). Gelatin, Bloom 300, was purchasedfrom United
StatesBiochemicalCorp. (Cleveland,OH).
Whole caseinwaspreparedfrom skim milk by acid precipitation.
Myosin was extracted from the pectoralis superficialisand pro-
fundus musclesof a freshly slaughteredmale chicken (broiler 11 -
12 wk old) with cold KCl-potassium phosphatebuffer, pH 6.5
accordingto the method of Perry (1955).
Preparation of protein samples
For hydrophobicity, solubility, emulsifying activity index,
emulsion stability index and fat binding capacity determinations,
the following protein samples were prepared. To increase the ac-
curacy of regression analysis, as many different extents of denatura-
tion as possible were attempted thus resulting in samples of differ-
ent solubility. For protein samples which needed constant pH to
reproduce the same degree of denaturation, the weakest phosphate
buffer (O.OlM) was used. For proteins in which the solubility was
difficult to change by heating, CaClz was used to disperse the pro-
teins.
Bovine serum albumin (BSA). A 1% solution in distilled water.
Heat denatured: the pH was adjusted to 4.0 and then the solution
was heated at 100°C for 5 min and homogenized in an Omni-Mixer
(Ivan Sorvall, Norwalk, CT) for 1 min at speed setting 1 (lowest
speed).
p-Lactoglobulin. A 1% solution in distilled water. Heat dena-
tured: the pH of the solution was adjusted to 1.0, heated at 100°C
for 15 min and then homogenized for 1 min at speed setting 1.
Soy protein isolate. A 1% aqueous dispersion, stirred on magnetic
stirrer for 5 min, and then, homogenized for 1 min at speed setting
5. Samples l-4: pH 5.5, heated at 100°C for 0.25,0.5,1.0, and 2.0
min, respectively. Sample 5: pH 5.5, autoclaved at 121°C for 15
min. Sample 6: pH 7.2, autoclaved at 121°C for 15 min.
Promine-D. A 1% dispersion in O.OlM phosphate buffer, pH 7.4,
stirred magnetically for 5 min and homogenized for 1 min at speed
setting 5.
Ovalbumin. A 1% aqueous solution. Samples l-4: pH 5.6,
heated at 80°C for 1.5, 2.0, 2.5, and 3.0 min, ;espectively.-Sample
5: pH 5.6, heated at 100°C for 5 min. Sample 6: pH 1.0, heated
at 100°C for 15 min.
Pea protein isolate (a). A 1% aqueous dispersion, stirred mag-
netically for 5 min, and then, homogenized for 1 min at setting 5.
Samples l-4: pH 5.8, heated at 80°C for 1, 2,4 and 7 min, respec-
tively, and then, homogenized for 10 sec.
Pea protein isolate (b). A 1% dispersion in O.OlM phosphate buf-
fer, pH 7.4, was magnetically stirred 5 min, and then, homogenized
for 1 min at speed setting 5.
Vital gluten. Samples l-5: 1% dispersions in 0.5, 1.0,1.5, 1.75
and 2.0 N acetic acid, respectively. Then, the samples were mag-
netically stirred for 5 mln, homogenized 15 set at speed setting 1,
and heated at 100°C for 30 min.
Canola protein isolate (a). A 1% aqueous dispersion, was stirred
for 5 min, and then, homogenized for 1 min at speed setting 5.
Samples l-4: pH 5.5, heated at 100°C for 0.5, 1.0, 1.5 and 2.0
min, respectively, and then homogenized for 5 set at speed setting
1. Sample 5: pH 7.2, autoclaved at 121°C for 10 min.
Canola protein isolate (b). A 1% dispersion in O.OlM phosphate
buffer, pH 7.4, was stirred for 5 min, and then, homogenized for
1 min at speed setting 5.
Sunflower protein isolate. A 1% dispersion in O.OlM phosphate
buffer, pH 7.4, was stirred for 5 min, and then, homogenized for
1 min at speed setting 5.
Whey protein. A 1% solution in 0.03M CaC12. Samples l-3:
pH 6.0, heated at 80°C for 4, 5 and 15 min, respectively. Sample
4: pH 5.8, heated at 80°C for 15 min.
Whole casein. A 1% solution in O.OlM CaClT, pH 7.4. Samples
1 and 2: 1% solution in O.OlM CaCl,, pH 7.4, a&claved at 12i’C
for 5 and 20 min, respectively. Sample 3: 1% solution in 0.02M
CaC12,pH 7.4, autoclaved at lil°C fo; 20 min.
Gelatin. A 1% dispersion in O.OlM phosphate buffer, pH 7.4,
stirred 5 min, and then, homogenized for 1 min at speed setting 5.
Heat denatured: 0.5% dispersion in O.OlM phosphate buffer, pH
7.4, heated at 75°C for 2.5 min with stirring.
Myosin. To form a stock solution for determination of hydro-
phobicity, solubility, and other functional properties, the original
gel was diluted twice with 0.3M NaCl in O.OlM phosphate buffer,
pH 7.4. The control sample was unheated and samples 1 and 2 were
heated at 75°C for 1 and 5 min, respectively.
Prior to analysis, all heated and unheated protein samples-
except Promine-D, pea protein (b), canola protein (b), sunflower
protein, gelatin, and myosin-were’ dialysed against O.OlM phos-
phate buffer, pH 7.4, containing 0.02% sodium azide. After hydro-
phobicity, solubility and emulsifying properties were determined, the
protein samples were freeze-dried and subsequently used for fat
binding capacity (FBC) determinations.
Protein (surface) hydrophobicity
Protein surface hydrophobicity was fluorometrically deter-
mined according to the method of Kato and Nakai (1980) after
slight modification. Each protein sample (2 ml) was serially diluted
with 0.01 M phosphate buffer, pH 7.4, to obtain protein concen-
trations ranging from 0.00156% to 0.05%. Two sets of protein
samples were prepared. Ten liters of cis-parinaric acid solution were
added to the first set of tubes. The parinaric acid-protein conjugate
was then excited at 325 nm and the relative fluorescence intensity
was measured at 420 nm in an Aminco-Bowman spectrofluoro-
meter, using a slit width of 0.5 mm. The fluorescence intensity read-
ing was standardized by adjusting the fluorometer reading to an
arbitrary value of 7.4 when 10 liters of cis-parinaric acid solution
was added to 2 ml of decane. Then, the fluorescence readings of the
protein samples were taken. The net fluorescence intensity at each
protein concentration was determined by subtracting the fluores-
cence intensity of each sample without cis-parinaric acid from the
fluorescence intensity of the corresponding sample containing
cis-parinaric acid. The initial slope (So) of the fluorescence intensity
vs protein concentration plot was used as an index of the protein
surface hydrophobicity. The initial slope was determined by linear
regression analysis using a Monroe (Orange, NJ) 1880 programmable
calculator.
Solubility index
Protein samples (l%, w/v, in O.OlM phosphate buffer pH 7.4)
were dispersed by stirring with a magnetic stirrer for 5 min and then
blended in a Sorval Omnimixer at speed setting 5 for 1 min. The pH
of each dispersion was adjusted to 7.4 by adding 1N NaOH. A por-
tion of each protein suspension was then centrifuged at 27,000 x g
for 30 min. Aliquots of the suspension and the supernatant after
centrifugation were diluted and the protein contents were deter-
mined by the Phenol-Biuret method (Brewer et al., 1974)..The per-
cent solubility index (s) was taken as the ratio of the protein con-
tent of the supernatant to that of the suspension.
Fat binding capacity (FBC)
The FBC was determined according to the method of Voutsinas
and Nakai (1981).
Emulsifying activity index (EAI)
EAI was determined by the tubidimetric technique of Pearse
and Kinsella (1978). To prepare emulsion, each protein sample
was diluted to a concentration of 0.5% with O.OlM phosphate buf-
fer, pH 7.4. Two ml of corn oil and 6 ml of the diluted protein dis-
persion were homogenized together in an Omni-mixer with a micro-
attachment (Ivan Sorvall, Inc., Norwalk, CT) at speed setting 1 for
1 min at 20°C.
Emulsion stability index (EN)
ES1 was determined by a modification of the method of Pearse
and Kinsella (1978) as follows: the emulsion, prepared as described
above (EAI), was held at room temperature and aliquots (0.1 ml)
were taken directly from the bottom of the container, containing
the emulsion, at different time intervals and diluted to 50 ml
(500 x dilution) with O.OlM phosphate buffer, pH 7.4, containing
0.1% sodium dodecyl sulfate (SDS). The absorbance of diluted
emulsions at 500 nm was then iecorhed with a Beckman DB spec-
trophotometer. The half-life (min) of the absorbance decay with
time, determined graphically, was used as ESI.
Viscosity
Viscosity measurements of 0.5% protein dispersion at 20°C were
made using a Brookfield Synchro-Lectric viscometer, Model LVT
fitted with a UL adapter, at 60 rpm.
Statistical analysis
Simple and multiple regression analyses were used to determine
the relationships between hydrophobicity, solubility and emulsify-
ing properties of the protein samples. These analyses were carried
out by using a Monroe 1880 programmable calculator. In addition,
backwards stepwise multiple regression analyses and surface visuali-
zation plotting were carried out using an Amdahl470 V/8 computer
at the University of British Columbia. Five independent variables
Volume 48 (1983hJOlJRNAL OF FOOD SCIENCE-27
 HYDROPHOBICITY AND EMULSIFYING PROPERTIES. ..

were used in the initial equation in the backwards stepwise proce- properties. It is evident that, for most proteins under a
dure including surface hydrophobicity (S,,), solubility index (s), given set of conditions, protein solubility decreased as heat-
interaction of So and s, and quadratic powers of So and s. Depend- ing time increased due to the progressive denaturation of
ent variables included EAI, ESI, and FBC. The models for the the protein. Gelatin, as expected, was completely solubil-
prediction of the emulsifying properties were selected on the basis ized on heating due to the rupture of hydrogen bonds
of the statistical s$nificance of F-probabilities of the partial regres-
sion coefficients. which are responsible for its insolubility (Blanshard, 1970).
As protein denaturation progressed, as seen by the decrease
RESULTS & DISCUSSION in protein solubility, the hydrophobicity usually increased.
This is due to the gradual exposure of hydrophobic amino
Effect of heat treatment on emulsifying properties acid residues of native proteins which are usually buried in
the interior of the molecules (Tanford, 1973) as a result of
Table 1 shows the relationships of hydrophobicity and the protein unfolding. In the case of whey protein, it can
solubility index of selected proteins with their emulsifying be seen, that excessive heating resulted in a decrease of

Table l-Relationships between protein hydrophobicity. solubility index, emulsifying activity, emulsion stability and fat binding capacity of
various proteinsa

Solubility
ESlb (min)
Hydrophobicity index (s/o) EAI
Protein EC-J) (s) (m*/g) I II FBC (%I

Bovine serum albumin -control 325 100.0 148 108.50 91.50 25.0
Bovine serum albumin - heated 304 26.8 140 90.00 109.00 54.4
P-Lactoglobulin -control 426 100.0 96 27.20 37.00 4.2
fi-Lactoglobulin - heated 192 6.4 51 25.30 27.50 16.5
Soy isolate - control 95 26.4 42 6.65 25.20 90.0
-sample 1 (pH 5.5, 100°C. 0.25 min) 97 26.4 51 7.30 26.00 86.3
-sample 2 (pH 5.5, 100°C. 0.5 min) 131 24.0 56 5.55 17.00 81.7
- sample 3 (pH 5.5, lOO’C, 1 .O min) 150 14.2 48 1 .oo 5.13 76.0
- sample 4 (pH 5.5, lOO’C, 2.0 min) 144 4.2 58 0.98 14.03 73.5
- sample 5 (pH 5.5, 121°C. 15 mini 160 8.2 50 0.76 11.20 68.4
. - sample6 (pH 7.2, 121’C. 15 min) 128 77.2 76 112.50 100.00 54.8
Promine-D 39 29.1 65 1.10 5.80 85.3
Ovalbumin -control 6 100.0 24 0.56 0.70 42.7
- sample 1 (pH 5.6, 80°C. 1.5 min) 38 90.6 34 0.63 1.10 45.9
- sample 2 (pH 5.6,BO”C. 2.0 mini 87 75.8 35 1.10 1.80 57.5
- sample 3 tpH 5.6,BO”C. 2.5 min) 95 70.6 45 1.08 1.70 66.9
- sample 4 (pH 5.6, BO’C, 3.0 min) 142 48.9 49 5.85 8.40 82.5
- sample 5 (pH 5.6, lOO’C, 5 min) 269 10.6 78 24.30 138.00 102.6
-sample6 (pH 1.0, lOO’C, 15min) 296 46.0 136 12.60 108.00 101.3
Pea isoiate (a) -control 66 42.6 61 3.67 15.50 66.4
- sample 1 (BO”C, 1 min) 77 34.7 66 0.60 0.42 53.9
- sample 2 (80°C. 2 min) 104 29.5 35 0.56 0.31 51.6
- sample 3 (80°C. 4 min) 100 25.2 37 0.85 0.16 50.4
-sample 4 (80°C. 7 min) 121 20.2 25 1.45 0.16 57.5
Pea isolate tb) 59 58.0 54 1.60 17.60 78.9
Gluten -sample 1 (0.5 N acetic, lOO”C, 30 min) 65 71.6 70 0.36 17.60 38.0
-sample 2 (1.0 N acetic, 100°C. 30 min) 59 79.0 69 0.55 25.00 41.8
-sample 3 (1.5 N acetic, lOO”C, 30 min) 65 85.6 82 0.28 23.80 34.8
- sample 4 (1.75 N acetic, lOO’C, 30 min) 60 89.6 74 0.80 26.60 39.9
-sample 5 (2.0 N acetic, lOO”C, 30 min) 57 93.3 69 0.78 23.30 32.2
Canola isolate (a) -control 65 28.3 60 0.54 3.76 33.9
- sample 1 tpH 5.5, 100°C. 0.5 min) 75 15.9 63 0.49 2.80 29.1
- sample 2 (pH 5.5, lOO”C, 1 .O min) 77 10.5 49 0.17 1.27 29.2
-sample 3 (pH 5.5, 100°C. 1.5 min) 86 3.6 41 0.12 0.45 35.1
-sample 4 (pH 5.5, 100°C. 2.0 min) 78 2.9 40 0.14 0.47 42.5
- sample 5 (pH 7.2, 121°C. 10 min) 101 21.2 50 2.20 8.40 35.7
Canola isolate (b) 55 44.0 66 0.25 3.35 56.0
Sunflower isolate 47 31 .o 75 1.40 5.30 105.8
Whey protein - control 182 88.7 87 50.30 87.50 74.5
- sample 1 (pH 6.0,8O”C, 4 min) 211 75.2 98 61.30 102.00 75.2
- sample 2 (pH 6.0,8O”C, 5 min) 164 65.5 89 63.00 104.00 72.5
- sample 3 IpH 6.0,80°C, 15 min) 128 50.4 82 48.30 104.50 100.8
- sample 4 (pH 5.8,8O”C, 15 min) 132 61.6 85 46.30 108.00 99.4
Casein - control 28 100.0 58 1.70 39.30 11.3
-sample 1 tO.OlM CaClg, 121°C. 5 min) 30 76.2 56 9.50 35.00 18.1
-sample 2 (0.01 M CaC12, 121°C. 20 min) 21 71.5 49 14.75 24.30 16.7
-sample 3 (0.02M CaClg, 121°C, 20 min) 23 70.2 50 12.50 16.80 14.0
Gelatin -control 5 15.3 46 4.65 3.10 19.1
-heated (75OC, 2.5 min) 6 100.0 59 10.20 9.40 __-_
Myosin -control 14= 1 oo.o= 43= 36.00’ _--_
- sample 1 (75°C. 1 min) 44c 50.9c 5oc 22.40’ ----
- sample 2 (75’C, 5 min) 54c 16.5’ 48’ 1o.ooc ---_

E average of duplicate determination

I: 0.1 M NaCl added; II: NaCl not added
’ 0.3 M NaCl added

28-JOURNAL OF FOOD SCIENCE-Volume 48 (7983)

hydrophobicity, due probably to participation of some emulsifying properties (EAI and ESI) of ovalbumin and
of the exposed hydrophobic groups in hydrophobic inter- lysozyme by heat denaturation was also reported by Kato
actions. For casein, heating did not result in any substan- and Nakai (1980) and Kato et al. (1981).
tial change of its hydrophobicity and this was expected, Another example of protein denaturation resulting in
since casein exists in a random coil conformation (Morr, improvement of functionality was reported by Aoki et al.
1979). The results in Table 1 also indicate that for samples (1981), who determined the effect of alcohol modifica-
having the same solubility, the more hydrophobic the tion of soy protein on its emulsion stabilizing properties.
protein, the greater are its emulsifying properties, The soy protein was denatured with 50% alcohol by treat-
Looking, specifically, at the results of soy protein in ment at 35OC for 2 hr. The emulsion stabilizing properties
Table 1, it can be seen that, the EAI of all heated samples of soy protein modified with ethanol or n-propanol de-
were slightly greater than that of the control. The ESI, creased with increasing solubility, whereas the emulsion
however, was initially slightly increased by heating, but stabilizing properties of the unmodified soy protein in-
subsequently started to decrease as the solubility progres- creased. Aoki et al. (1981) attributed the improved emul-
sively decreased; that is, solubility became an increasingly sion stability brought about by the alcohol modification of
important factor controlling this property. The higher EAI soy protein to the perturbation and unfolding of the hy-
values of heated soy protein samples were probably due to drophobic interior structure of the native soy protein by
their increased hydrophobicity values. On the other hand, alcohol, and to the resulting increase in the exposed hydro-
the observed decrease in ES1 upon heating may be due to phobic amino acid residues. It should be noted, that the
a decreasein solubility as a result of aggregation. increased emulsion stabilizing properties of soy protein be-
The proteins studied here can be divided into four tween pH 2 and 7, achieved through alcohol modification
groups according to the effect of heating on their emulsi- by Aoki et al. (19 8 1), is very important because soy protein
fying properties. The first group includes BSA, gluten, and can be expected to play a significant role in the stabiliza-
whey protein whose EAI was not substantially affected tion of a wide range of food emulsions, all meat emul-
by heating, whereas ES1 was improved by heating. The sions falling well within these pH limits.
second group includes soy protein and myosin whose EAI
was slightly improved by heating, whereas the ES1 was Statistical analysis
decreased. The third group includes @-lactoglobulin, pea, Regression equations for predicting the emulsifying and
canola, and casein, whose emulsifying properties were ad- fat binding properties from hydrophobicity and solubility
versely affected by heating. Finally the fourth group in-
data of heat denatured proteins in Table 1 are presented in
cludes ovalbumin and gelatin whose emulsifying properties
Table 3. Simple linear regression analyses showed that the
were markedly improved upon heating. It is evident, there-
coefficients of determination (r*) between S, and EAI,
fore, that heating does not have uniform effects on the
and between Se and ES1 (detetmined without NaCl) were
emulsifying properties of different proteins.
highly significant (P < 0.001). No significant correlation
The improvement of emulsifying properties of gelatin was found, however, between solubility and EAI or ESI.
upon heating might have been due to the increase in solu-
bility since its hydrophobicity was not changed by heat- Moreover, no correlation was found between So or solu-
bility and FBC. Although significant correlations were
ing. Kato and Nakai (1980), and Kato et al. (1981) re- found between S, and ES1 as well as solubility and ES1
ported that the emulsifying properties (EAI and ESI) of
determined in the presence of O.lM NaCl, such correla-
ovalbumin and lysozyme were markedly improved upon tions are not considered meaningful since the effect of
heating. They also found that this improvement was cor-
O.lM NaCl on solubility and S, were not determined. It
related with the higher hydrophobicity of heat denatured
is reasonable that 0.1 M NaCl may positively or negatively
protein samples as compared to those of the native pro-
affect the solubility of a protein. Sodium chloride was
teins. Furthermore, their results indicated that the more
used in the ES1 test, to obtain a general idea on the salt
hydrophobic the proteins, the greater the decreasein inter-
sensitivity of the emulsions formed by different proteins.
facial tensions and the improvement in emulsifying proper-
ties. During the handling of the ovalbumin samples we Relatively high salt concentrations are used in some foods,
e.g. meat products. As seen in Table 1, O.lM NaCl usually
noticed that the apparent viscosities of heat-denatured
has a detrimental effect upon the emulsion stability. NaCl
samples increased. This is evident in Table 2, which shows
exerts its negative effect on emulsion stability probably by
the apparent viscosity changes of some proteins upon heat-
ing. It is suggested, therefore, that higher viscosity appears
to be another factor contributing to the great improvement Table 2 - Effect of heating on the apparent viscosity of some
of emulsifying properties of heat denatured ovalbumin proteinsa measured at 20°C
samples. Apparent viscosity
As shown in Table 1, the FBC of different proteins is Proteina 1Pa.s x 1 O-3)
differently affected by heating. While the FBC of Canola
Whey - control 1.14
protein is relatively unaffected by heating, the FBC’s of - sample 1 IpH 6.0, 80°C, 4 min) 1 .I4
BSA, /I-lactoglobulin, ovalbumin, whey protein, and casein - sample 2 (pH 6.0, 80°C, 5 min) 1.14
are positively affected, and the FBC’s of soy and pea pro- - sample 3 (pH 6.0, 80°C, 15 min) 1.16
teins are adversely affected by heating, - sample 4 (pH 5.8, 80°C, 15 min) 1.20
Protein denaturation usually decreases solubility then
adversely affects protein functionality. However, as shown BAS - control 1.13
in this study, emulsifying and fat binding properties of - heated (pH 4.0, IOO’C, 5 min) 1.24
some proteins (ovalbumin, whey protein) can be improved
by denaturation. According to Morr (1979) denaturation of Ovalbumin - control 1 .I0
- sample 1 (pH 5.6,80°C, 1.5 min) 1 .I3
the whey protein molecule, if produced at the proper stage - sample 2 (pH 5.6, 80°C, 2.0 min) 1.13
of the protein concentrate isolation/utilization process, can - sample 3 (pH 5.6, 8O’C. 2.5 min) 1.16
improve the functionality. The improvement in functional- - sample 4 (pH 5.6,80°C, 3.0 min) 1 .I 7
ity is probably due to an unfolding of the molecule to ex- - sample 5 (pH 5.6,lOO”C. 5 min) 1 .I8
pose hydrophogic amino acid residues, thus making the - sample 6 (pH 1 .O, 100°C. 15 min) 1.90
protein more amphiphilic and capable of orienting at the
oil-water interface (Morr, 1979). A great improvement of a0.5% protein in 0.01 M phosphate buffer, pH 7.4

Volume 48 /1983)-JOURNAL OF FOOD SCIENCE-29

 I HYDROPHOBICITY AND EMULSIFYING PROPERTIES. . .

reducing the surface charge and by withdrawing surface
water of oil droplets.
As seen in Table 3, multiple regression analysesbetween
S,, solubility and EAI or ESI, show highly significant corre-
lations (R* values are 0.542 and 0.434, P < 0.001, respec-
tively).
Since the Student’s t-values of partial regression coeffi-
cients are highly significant, both hydrophobicity and solu-
bility significantly affect the emulsifying properties of the
proteins. As shown in Table 3, the backwards stepwise
multiple regression analysis gives a highly significant corre-
lation between S,, solubility and FBC of the 48 protein
samples in Table 1. Comparing the regression models in
Table 3, it is apparent that the models obtained by step-
wise regression analysis have higher R* values and lower
S.E. (standard error of the estimate) values than those of
the linear regression models. Our discussion, therefore,
will be confined within the models calculated by stepwise
regression analysis.
The EAI of the proteins studied was significantly affected
by So, solubility and the square of solubility (Table 3).
The R* value for EAI was 0.583, indicating that 58.3% of
the variability in EAI of the protein studied could be
accounted for by the three independent variables S,, s
.and s*. In this model, the statistical significance of the
square of solubility index indicates that as the value of solu-
bility index increases, the effect of that variable declines,
i.e., the EAI response to increasing levels of solubility may
be depicted as curvilinear rather than linear. Response
surface plots (contours) were generated by the computer
to visualize the effects of S6 and solubility on the function-
al properties studied. As shown in Fig. 1, regardless of the
solubility, as hydrophobicity increases the EAI is initially
increased and then slightly declined. Although at low and
medium hydrophobicity values, increasing solubility levels
increase the EAI, at very high S, values, solubility does not
appear to play an important role in the emulsifying capac-
ity of proteins.
The ES1 of the proteins is significantly affected by So, s,
the interaction of S, and s, and the s2 (Table 3). The R*
value of the model is 0.584, indicating that 58.4% of the
variability in ES1 could be accounted for by these 4 vari-
ables. As shown in Fig. 2, regardless of the solublity, as
S, increases the ES1 is initially increased and then de-
creased. At low and medium So values, increasing solubility
index levels increases ESI. However, at very high S,, values,
solubility index does not appear to be important for the
emulsion stability.
The FBC of the proteins studied was significantly af-
fected by Se, s, and the squares of these two variables
(Table 3). The R* of the model was 0.473 (P < O.OOl),
indicating that 47.3% of the variability in FBC could be
accounted for by the four variables. The statistical signifi-
cance of quadratic powers of S, and solubility in the FBC
model in Table 3 indicates that as the value of S, or s
increases, the FBC value increases first then declines (i.e.
the response of FBC to increasing levels of S, or solubil-
ity can be depicted as a curvilinear graph).
The finding of this study that with increasing S, the
emulsifying properties initially increase and then decrease
can be explained by taking into account the fact that the
emulsifying properties of proteins ultimately depend on the

Table 3 - Regression models for prediction of emulsifying and fat binding properties of heat denatured proteins.
I
Dependent Regression Variable Regression
variable analysis description coefficient F-ratio Fprobability t-valuea p-value

EAI Simple Constnat 41.162

(n=52; r2=0.464, linear so 0.202
P<O.OOl ;
S.E.b=18.95)

EAI Multiple Constant 29.283

(n=52: R2~0.542. linear so 0.207 - 7.202X” 0.698
P<O.OOl ; S 0.226 - 2.887” 0.280
s.E.b=I 7.78)

EAI Backwards Constant 16.877 4.531 0.039

(n=52; R2~0.583, multiple SO 0.214 58.77 0.000 0.720
P<O.OOI ; nonlinear S 0.932 7.83 0.008 1.151
s.E.b=17.401 s2 -0.007 4.73 0.035 - -0.894

ESI Simple Constant 0.880

(n=49; r2=0.377, linear so 0.274
P<0.001;
S.E.b=30.98)

ESI Multiple Constant -14.308

(n=49; R2=0.434, linear so 0.278 5.613*** 0.623
P<O.OOl ; S 0.295 2.153 0.239
s.E.b=29.53)

ESI Backwards Constant -69.463 18.02 0.000

(n=49; R2=0.584, multiple so 0.565 34.31 0.000 1.268
P<O.OOI ; nonlinear 2.034 13.62 0.001 - I.651
S.E.b=26.70) i xs -0.004 10.08 0.003 - -0.779
52 -0.012 5.92 0.019 -1.044

FBC Backwards Constant 4.895 0.16 0.691 -

(n=48; R2=0.473. multiple so 0.451 13.84 0.001 I .445
P<O.OOl ; nonlinear s 1.398 13.84 0.001 I .602
S.E.b=20.97) sg -0.001 10.15 0.003 - -1.270
S2 -0.014 11.79 0.001 - -1 .J45

a*: ~03.05; **: P<O.OI; ***: P<O.OOI

btandard error of estimate

30-JOURNAL OF FOOD SCIENCE-Volume 48 (1983)

suitable balance between the hydrophile and lipophile, and
do not necessarily increase as the proteins become more
lipophilic (Aoki et al., 198 1). Excessive denaturation of the
soy protein by n-propanol resulted in a lower emulsion sta-
bility as compared to the moderate denaturation by ethanol
(Aoki et al. 1981). There may be many factors, e.g., molec-
ular size, molecular flexibility and charge, besides the
balance of hydrophile and lipophile which participate in
determining the emulsifying properties of proteins.
Wolf and Cowan (1975) reported that in ground meat
products fat absorption or binding appeared to be involved
in formation and stabilization of an emulsion. Thus, they
suggested that fat adsorption may simply be another
aspect of emulsification. This helps to explain the obser-
vation in this study that hydrophobicity has a curvilinear
effect also on FBC, as the concept of hydrophile-lipophile
balance can be utilized again. Voutsinas and Nakai (1981)
already reported that with increasing S, the FBC was ini-
tially increased and then decreased.
Pearson et al. (1965) reported that only that fraction of
protein which is soluble can function as an effective emulsi-
fying agent. hloreover, Franzen and Kinsella (1976) sug-
gested that, as a protein becomes more soluble, it forms
layers around the fat droplet to facilitate association with
the aqueous phase. Granular, insoluble proteins, however,
separate from the oil phase or just float on the oil surface
where they remain inert and contribute little toward emul-
sification. Similarly, soluble proteins enclose the fat glob-
ules and render the emulsion more stable to heat treatment.
Atso,, according to Bull (1972), the surface activity of a
protein is a function of the ease with which the protein can
52 65 79,
9 78 148
Fig. I-Emulsifying activity index response surface
7
9 78 148
Fig. %Emukion stability index response surface
migrate to, adsorb at, unfold, and rearrange at an interface.
Therefore, solubility in the aqueous phase, is closely re-
lated to surface activity of proteins (Kinsella, 1979). Many
authors, on the other hand, have reported that high levels
of solubility were not necessarily associated with maximum
emulsifying properties (Aoki et al., 1980; McWatters and
Cherry, 1975; McWatters and Holmes, 1979a, 1979b;
Smith et al., 1973; Wang and Kinsella, 1976). Flint and
Johnson (1981) investigated a film formation by soy pro-
tein (isolate) at an oil-water interface for the pH range
l-10. Definite film formation was observed at all pH
values below the isoelectric point (4.6) of the protein
and up to pH 6.5. At pH 5.4, despite the low solubility
strong film formation occurred. However, at pH’s above 5.4
the strength of films formed gradually decreased until at an
upper limit pH of 7.5, the presence of an interfacial layer
could barely be seen. The marked pH dependence of
film formation on the alkaline side of the isoelectric point
was attributed to the fact that with increasing pH the pro-
tein becomes more soluble in the aqueous phase and con-
sequently less likely to be brought out of solution (coagu-
lated) at the phase boundary. The ability of soy protein to
form a film even at very low pH’s, where the solubility of
protein is high, suggestedthat this phenomenon may not be
linked solely to solubility but also to the availability of
lipophilic groups for binding at the oil-water interface.
Changes in the protein conformation may occur at acid
pH’s which enhance the combination of the protein and oil
leading to the formation of interfacial film described
(Flint and Johnson, 1981). -Continued on nextpage
_
217 286 355 425
contour as a function of hydrophobicity (So) and solubility index (s).
217 286 355 425
SO
contour as a function of hydrophobiciv (So) and solubijiw index (s).
Volume 48 (1983)-JOURNAL OF FOOD SCIENCE-31
 HYDROPHOBICITY AND EMULSIFYING PROPERTIES. ..

Considering the above contradictory situation, the re- McWatters, K.H. and Cherry, J.P. 1975. Functional properties of
ueanut paste as affected by moist heat treatment of full-fat pea-
sults of this study showed that high solubility had a nega- nuts. J. Food Sci. 40: 1205..
tive effect on FBC in agreement with the results of Vout- McWatters. K.H. and Holmes, M.R. 1979a. Influence of pH and salt
sinas and Nakai (1982). They attributed the low FBC of concentration on nitrogen solubility and emulsification properties
of soy flour. J. Food Sci. 44: 770.
soluble proteins to their conformation (mainly helical) McWatters, K.H. and Holmes, M.R. 1979b. Influence of moist heat
which may not permit their binding sites to be sterically on solubility and emulsification properties of soy and peanut
flours. J. Food Sci. 44: 774.
available for interaction with oil or to the limited accessof Morr, C.V. 1979. Conformation and functionality of milk proteins.
oil (hydrocarbon chains) to the protein binding sites due to In “Functionality and Protein Structure,” Ed. A. Pour-El, Amer.
a possible barrier around them formed by the excessive Chem. Sot. SvmDosium Series 92. Washington. DC.
Nakai, S., Ho,-L.; Tung, M.A., aid Quinn, J.g. 1980a. Solubiliza-
number of protein polar groups. tion of rapeseed, SOY and sunflower protein isolates by surfactant
In conclusion, the emulsifying properties of proteins are and proteinase treatments. Can. Inst. Food Sci. Technol. J. 13: 14.
Nakai. S., Ho, L., Helbig. N., Kato, A.. and Tung, M.A. 1980b.
influenced by both hydrophobicity and solubility. Al- Relationship between hydrophobicity and emulsifying properties
though important, solubility alone cannot fully explain the of some plant proteins. Can. Inst. Food Sci. Technol. J. 13: 23.
Nakai, S. and Powrie, W.D. 1981. Modification of proteins for func-
emulsifying properties especially when the proteins are tional and nutritional improvements. In “Cereals: A Renewable
heat denatured. Resource ” Ed. Y. Pomeranz and L. Munck, Amer. Assoc. Cereal
Chem., S;. Paul, MN.
Pearse, K.N. and Kinsella, J.E. 1978. Emulsifying properties of pro-
REFERENCES teins: evaluation of a turbidimetric technique. J. Agric. Food
Chem. 26: 716.
Aoki, H., Taneyama, 0.. and Inami, M. 1980. Emulsifying proper- Pearson, A.M., Spooner, M.E., Hegarty, G.R.. and Brat&r, L.J.
ties of SOY protein: Characteristics of 7s and 11s proteins. J. 1965. Emulsifying capacity and stability of soy sodium pro-
Food Sci. 45: 534. teinate, potassium caseinate and non-fat dry milk. Food Tech-
Aoki, H., Taneyama, 0.. Orimo, N.. and Kitagawa, I. 1981. Effect nol. 19: 1841.
of lipophilization of soy protein on its emulsion stabilizing proper- Perry. S.V. 1955. Myosin adenosinetriphosphatase. In “Methods
ties. J. Food Sci. 46: 1192. in Enzymology,” Ed. S.P. Colowick and N.O. Kaplan, Vol. 2.
Blanshard, J.M.V. 1970. Stabilizers -their structure and properties. Academic Press, New York, NY.
J. Sci. Food Agric. 21: 393. Smith, G.C., Juhn, H., Carpenter, Z.L., Mattil, K.F., and Carter,
Brewer, J.M., Pesce, A.J., and Ashworth, R.B. 1974. “Experimental C.M. 1973. Efficacy of protein additives of emulsion stabilizers in
Techniques in Biochemistry.” Prentice-Hall, Inc.. Englewood frankfurters. J. Food Sci. 38: 849.
Cliffs. NJ. Swift, C.E. and Sulzbacher, W.L. 1963. Cornminuted meat emul-
Bull. H.B. 1972. Adsorbed surface films of egg albumin. J. Colloid sions: factors affecting meat proteins as emulsion stabilizers.
Interfacial Sci. 41: 305. Food Technol. 17: 224.
Crenwelge, D.D., Dill, C.W., Tybor, P.T.. and Landmann, W.A. Tanford, C. 1973. “The Hydrophobic Effect: Formation of Mice&s
1974. A comparison of the emulsification capacities of some pro- and Biological Membranes.” John Wiley & Sons, New York, NY.
tein concentrates. J. Food Sci. 39: 175. Volkert, M.A. and Klein, B.P. 1979. Protein dispersibility and emul-
Flint, F.O. and Johnson, R.F.P. 1981. A study of film formation sion characteristics of flour soy products. J. Food Sci. 44: 93.
by soy protein isolate. J. Food Sci. 46: 1351. Voutsinas. L.P. and Nakai, S. 1982. A simple turbidimetric method
Franzen, K.L. and Kinsella, J.E. 1976. Functional properties of suc- for determining the fat binding capacity of proteins. J. Agric.
cinylated and acetylated leaf protein. J. A&c. Food Chem. Food Chem. (In press).
24: 914. Wang,-J.C. and Kinselia, J.E. 1976. Functional properties of novel
Johnson, D.W. 1970. Functional properties of oilseed proteins. J. proteins: alfalfa leaf proteins. J. Food Sci. 41: 286.
Amer. Oil Chem. Sot. 47: 402. Wolf. W.J. and Cowan. J.C. 1975. “Sovbeans as a Food Source.”
Kato, A. and Nakai. S. 1980. Hydrophobicity determined by a CR’C Press, Inc., Clewland, OH. -
fluorescence probe method and its correlation with surface proper- Yasumatsu, K., Sawada, K., Moritaka, S., Misaki, M., Toda, J..
ties of proteins. Biochim. Biphys. Acta. 624: 13. Wada, T., and Ishii. K. 1972. Whipping and emulsifying properties
Kate, A., Tsutsui, N., Kobayashi, K., and Nakai, S. 1981. Effects of of soybean products. Agric. Biol. Chem. 36: 719.
partial denaturation on surface properties of ovalbumin and Ms received 5117182; revised 8/g/82; accepted g/6/82.
lysozyme. Agr. Biol. Chem. 45: 2755.
Keshavarz, E. and Nakai. S. 1979. The relationship between hydro- A part of this work was presented at the 41st Annual Meeting of
phobicity and interfacial tension of proteins. Biochim. Biophys. the Institute of Food Technologists, Atlanta, GA, June 7-10, 1981.
Acta. 575: 269. This research was conducted according to the terms of contract
Kinsella, J.E. 1976. Functional properties of proteins in foods: a no. OSU80-00026 for Agriculture Canada. (Scientific authority:
survey. Crit. Rev. Food Sci. Nutr. 7: 219. Dr. V. Harwalkar). The authors are grateful for the encouragement
Kinsella, J.E. 1979. Functional properties of soy proteins. J. Amer. given by late Dr. J.R. Quinn.
Oil Chem. Sot. 56: 242.

ELECTRICAL STIMULATION OF RABBITMUSCLE.. . From page 22

Locker, R.H., Davey, C.L., Nottingham, P.M.. Haughey, D.P., and trical stimulation on beef, pork. lamb and goat meat. Proc. 26th
Law, N.H. 1975. New concepts in meat processing. Adv. Food Eur. Meet. Meat Res. Work., Colorado Springs, CO, P. 19.
Res. 21: 157. Spudich, J.A. and Watt, S. 1971. The regulation of rabbit skeletal
Lowey, S. and Cohen, C. 1962. Studies on the structure of myosin. muscle contraction. J. Biol. Chem. 246: 4866.
J. Mol. Biol. 4: 293. Taylor, A.A., Shaw, B.G. and MacDougall, D.B. 1980. Hot debonlng
Mannherz, H.G., Brehme, H.. and Lamp, U. 1975. Depolymerization beef with and without electrical stimulation. Meat Sci. 5: 109.
of F-actin to G-actin and its repolymerization in the presence of Tonomura, Y., Tokura, S., Sekiya, K., and Imamura, K. 1961.
analogs of adenosine triphosphate. Eur. J. Biochem. 60: 109. Influence of solvent composition on the molecular shape and
Olson. D.G., Parrish, F.C., Dayton, W.R.. and Gall, D.E. 1977. Effect enzymic activity of myosin A. Arch. Biochem. Biophys. 95: 229.
of postmortem storage and calcium activated factor on the myo- Weber, K. and Osborn, M. 1969. The reliability of molecular weight
fibrillar proteins of bovine skeletal muscle. J. Food Sci. 42: 117. determinations by dodecyl sulfate-polyacrylamide gel electro-
Rizzino, A.A.. Barouch. W.W., Eisenberg. E. and Moos, C. 1970. phoresis. J. Biol. Chem. 244: 4406.
Actin-heavy meromyosin binding: determination of binding West, R.L. 1979. Effects of high temperature conditioning on
stoichiometry from ATPase kinetic measurements. Biochemistry muscle tissue. Food Technol. 33(4): 41.
9: 2402. Whiting, R.C., Strange, E.D., Miller, A.J., Benedict. R.C., Mozersky,
Savell, J.W., Smith, G.C.. Dutson, T.R., Carpenter, Z.L., and Suter. S.M., and Swift, C.E. 1981. Effects of electrical stimulation on the
D.A. 1977. Effect of electrical stimulation on palatability of beef. functional properties of lamb muscle. J. Food Sci. 46: 484.
lamb and goat meat. J. Food Sci. 42: 702. Will, P.A., Henrickson, R.L., Morrison, R.D., and Odell, G.V. 1979.
Shaw. F.D. and Walker, D.J. 1977. Effect of low voltage stimulation Effect of electrical stimulation on ATP depletion and sarcomere
of beef carcasses on muscle PH. J. Food Sci. 42: 1140. length in delay-chilled bovine muscle. J. Food Sci. 44: 1646.
Smith, G.C., Savell, J.W., Dutson. T.R., Hostetler, R.L., Terrel, MS received 6112182; revised g/30/82; accepted 10/6/82.
R.N., Murphey. C.E., and Carpenter. Z.L. 1980. Effects of elec-

~&JOURNAL OF FOOD SCIENCE-Volume 48 (1983)