Food Hydrocolloids 30 (2013) 647e655

Contents lists available at SciVerse ScienceDirect

Food Hydrocolloids
journal homepage: www.elsevier.com/locate/foodhyd

Effects of ultrasound on structural and physical properties of soy protein isolate

(SPI) dispersions
Hao Hu a, Jiahui Wu a, Eunice C.Y. Li-Chan b, Le Zhu a, Fang Zhang a, Xiaoyun Xu a, Gang Fan a,
Lufeng Wang a, Xingjian Huang a, Siyi Pan a, *
a
College of Food Science and Technology, Huazhong Agricultural University, No. 1 Shizishan Road, Wuhan, Hubei 430070, PR China
b
The University of British Columbia, Faculty of Land and Food Systems, Food Nutrition and Health Program, 2205 East Mall, Vancouver, British Columbia, Canada V6T 1Z4

a r t i c l e i n f o a b s t r a c t

Article history: The effects of low-frequency (20 kHz) ultrasonication at varying power (200, 400 or 600 W) and time (15
Received 10 May 2012 or 30 min) on functional and structural properties of reconstituted soy protein isolate (SPI) dispersions
Accepted 6 August 2012 were examined. Ultrasonic treatments reduced both the storage modulus and loss modulus of SPI
dispersions and formed more viscous SPI dispersions (fluid character). Moreover, ultrasound treatment
Keywords: significantly decreased the consistency coefficients and increased the flow behaviour index of SPI
Ultrasound
dispersions. Scanning electron microscopy of lyophilized ultrasonicated SPI showed different micro-
Soy protein isolate
structure with larger aggregates compared to non-treated SPI. No significant change was observed in the
Rheological property
Secondary structure
protein electrophoretic patterns by SDS-PAGE. However, free sulfhydryl content, surface hydrophobicity
Free sulfhydryl content and protein solubility of SPI dispersions were all increased with ultrasonic treatment. Differences in
Surface hydrophobicity solubility profiles in the presence versus absence of denaturing (0.5% sodium dodecyl sulphate and 6 M
urea) and reducing (mercaptoethanol) agents suggested a decrease in non-covalent interactions of SPI in
dispersion after ultrasonic treatment. Secondary structure analysis by circular dichroism indicated lower
a-helix and random coil in SPI treated at lower power, in contrast to higher a-helix and lower b-sheet in
SPI treated with higher power (600 W). In conclusion, under the conditions investigated in this study,
ultrasonic treatment resulted in partial unfolding and reduction of intermolecular interactions as
demonstrated by increases in free sulfhydryl groups and surface hydrophobicity, leading to improved
solubility and fluid character of SPI dispersions, while larger aggregates of ultrasonic-treated SPI in the
dry state were formed after lyophilization.
Crown Copyright Ó 2012 Published by Elsevier Ltd. All rights reserved.

1. Introduction be used to physically or chemically alter food properties. The effects

of ultrasound on liquid systems are mainly due to the cavitation
Application of ultrasound technology in food industry is phenomenon (Soria & Villamiel, 2010). Cavitation bubbles are
currently attracting much attention (Jambrak, Lelas, Mason, Kresic, formed during sonication (Gulseren, Guzey, Bruce, & Weiss, 2007).
& Badanjak, 2009). Ultrasound technology is based on mechanical The bubbles are rapidly formed and violently collapse, leading to
waves at a frequency above the threshold of human hearing extreme temperatures (5000 K) and pressures (1000 atm) which
(>16 kHz), and can be divided into two frequency ranges, that is can produce very high shear energy waves and turbulence in the
high frequency (100 kHze1 MHz, power < 1 W cm
2) and low- cavitation zone. The above-mentioned factors of temperature,
frequency (16e100 kHz, power in the range 10e1000 W cm
2) pressure and turbulence combine together to affect the ultrasound
(Soria & Villamiel, 2010). Low intensity (high frequency) ultra- treated system (Soria et al., 2010). Moreover, ultrasound can
sound is most commonly applied as an analytical technique to generate highly reactive free radicals from water molecules
provide information on the physicochemical properties of food such (H2O / H þ OH), leading to reactions with other molecules.
as firmness, ripeness, sugar content, acidity, etc (Soria & Villamiel, The effect of high-intensity ultrasound (HUS) on protein modi-
2010). In contrast, high-intensity (low-frequency) ultrasound can fication has been the subject of much research in recent years.
Some of the research studies have investigated the application of
HUS during protein chemical reaction or as pretreatment, so as to
* Corresponding author. Tel.: þ86 027 87283778; fax: þ86 027 87288373. promote the subsequent modification. For instance, Mu et al. (2010)
E-mail addresses: siyipan@yahoo.com.cn, pansiyi@mail.hzau.edu.cn (S. Pan). reported that ultrasound treatment was an efficient method for

0268-005X/$ e see front matter Crown Copyright Ó 2012 Published by Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.foodhyd.2012.08.001
648 H. Hu et al. / Food Hydrocolloids 30 (2013) 647e655

forming protein and polysaccharide conjugates, while Chen, Chen, Sinopharm Chemical Reagent Co., Ltd, Shanghai, China. All reagents
Ren, and Zhao (2011a) reported on the use of ultrasound were analytical grade.
pretreatment to increase the enzymatic hydrolysis of protein. Other
studies have applied HUS to directly modify the functional and 2.2. Ultrasound treatment of samples
physical properties of proteins, such as solubility, gelation, emul-
sification and foamability. For example, Madadlou, Emam-Djomeh, SPI dispersions (10.0%, w/v) were prepared by adding SPI
Mousavi, Mohamadifar, and Ehsani (2010) found that ultrasonic powder into distilled water and then gently stirring overnight at
treatment of casein could postpone the gelation point and increase ambient temperature. An ultrasound processor (NingBo Scientz
the firmness of casein gel, while Madadlou, Mousavi, Emam- Biotechnology Co. Ltd., Ningbo, China) with a 0.636 cm diameter
Djomeh, Ehsani, and Sheehan (2009) achieved a decrease in the titanium probe was used to sonicate 100 mL of SPI dispersions in
turbidity of casein solutions by ultrasonication. Kresi c, Lelas, 100 mL flat bottom conical flasks which were immersed in an
Jambrak, Herceg, and Brncic (2008) reported that ultrasonication ice-water bath. Samples were treated at 20 kHz at different levels
of whey proteins significantly increased their solubility and of power output (0 W, 200 W, 400 W, 600 W) for 15 min
apparent viscosity, and altered their flow behaviour to be shear- and 30 min (pulse duration of on-time 5 s and off-time 1 s) After
thickening. Solubility and foaming ability of whey proteins ultrasound treatment all samples were lyophilized and
(Jambrak, Mason, Lelas, Herceg, & Herceg, 2008) and a-lactalbumin then stored at room temperature in air tight containers until
(Jambrak, Mason, Lelas, & Kresic, 2010) were improved by HUS. analyzed.
Similarly, HUS treatment significantly changed conductivity and
rheological properties, and increased solubility, specific surface 2.3. Determination of ultrasound power and intensity
area and emulsifying activity index values of soy protein concen-
trates (Jambrak et al., 2009), as well as the protein solubility and Ultrasonic power, which is considered as mechanical energy,
gelling ability of commercial soy protein isolate (SPI) (Tang, Wang, would be partially lost in the form of heat when ultrasound passes
Yang, & Li, 2009). Zisu, Bhaskaracharya, Kentish, and Ashokkumar through the medium (Jambrak et al., 2009). Since the ultrasonic
(2010) reported that pilot scale sonication reactors operating at irradiation of a liquid produces heat, recording the temperature as
a frequency of 20 kHz could be used to reduce the viscosity and to a function of time leads to the acoustic power estimation (in W) by
improve gelling of whey and casein containing dairy ingredients. the equation P ¼ m$cp$(dT/dt), where m is the mass of the sonicated
Thus HUS not only represents a rapid, efficient and reliable alter- liquid (g), cp is specific heat of the medium at a constant pressure
native to improve the food protein quality, but it also has the (J g
1 K
1) and is dependent on composition and volume of the
potential to develop new products in food industry (Soria & medium, and dT/dt is the slope at the origin of the curve. Ultrasonic
Villamiel, 2010). intensity was measured by calorimetry using a thermocouple
Some studies have investigated the molecular structural (model: TASI 601, TASI Ltd., Suzhou, China) and expressed in
changes after HUS treatment. Increased intra-molecular mobility W cm
2.
and surface activity Guzey, Gulseren, Bruce, and Weiss (2006) and Using ultrasonic treatment with the 20 kHz probe at power
changes in the free sulfhydryl groups, particle size, surface hydro- output of 200 W, 400 W and 600 W, the ultrasonic intensity was
phobicity, and secondary structure (Gulseren et al., 2007) were 75e83 W cm
2, 105e110 W cm
2 and 131e138 W cm
2,
reported to result from ultrasonically induced changes in the respectively.
molecular structure of bovine serum albumin. Kresi c et al. (2008)
reported that sonication did not alter the thiol content but resul- 2.4. Rheological measurement
ted in minor changes to the secondary structure and hydropho-
bicity of whey proteins. Rheological measurements were carried out using a controlled
Soy protein isolate is a commercial soy protein product having at stress rheometer, according to the method of Sittikijyothin,
least 90% protein (dry basis) which has been widely applied in the Sampaio, and Gonçalves (2010) with some modifications as
food industry as an important ingredient due to its nutritional described below. Protein dispersions (12.5%, w/v) of lyophilized SPI
value, desirable functional properties and low cost (Chen, Chen, were prepared in distilled water. All rheological measurements
Ren, & Zhao, 2011b; Wang et al., 2008). However, to the best of were performed at 25  C using a controlled stress rheometer
our knowledge, little is known about the effects of varying the AR2000 (TA Instruments, Leatherhead, UK) fitted with parallel
conditions of HUS treatment on the structure of SPI proteins. Thus, plates (40 mm diameter and 1 mm gap). The following sequence of
the purpose of this study was to investigate the effects of HUS measurements was done:
treatment on the functional and structural properties of SPI, as
a function of the ultrasound power (200e600 W) and time (15e 2.4.1. Strain sweep analyses
30 min) of treatment, in order to obtain a better understanding of Stress sweep analyses were performed in order to determine
physicochemical effects of HUS on SPI which may lead to improving the linear viscoelastic region (LVR) of each dispersion. In
its applications in the food industry. a typical experiment, the sample was placed on the plate and
the stress sweep test at a frequency of 1 Hz was recorded. The
2. Materials and methods limit of linear viscoelastic region was determined using the
criterion adopted by Caillard, Remondetto, and Subirade (2010).
2.1. Materials The limit of viscoelastic response is represented as the point
beyond which the storage modulus (G0 ) begins to deviate by 5%
Soy protein isolate (SPI, >90% protein, as determined by the of its maximum value. All experiments were performed in
Kjeldahl method) and 5, 50 -dithiobis-(2-nitrobenzoic acid) (DTNB) duplicate.
were purchased from Aladdin Reagent Company, Shanghai, China.
Tris base was obtained from the Dow Chemical Company, USA. 2.4.2. Frequency sweep analyses
Glycine, sodium dodecyl sulphate (SDS) and 1-anilino-8- Frequency sweep analyses were performed for a frequency
naphthalene-sulfonate (ANS) were obtained from Sigma Chemical range of 0.1e100 rad s
1. The values used for the strain amplitude
Co., St. Louis, MO, USA. Other reagents were obtained from depended on the strain sweep analyses (Section 2.4.1).
 H. Hu et al. / Food Hydrocolloids 30 (2013) 647e655 649

2.4.3. Flow curve analyses was serially diluted with the same buffer to obtain protein
Flow curve analyses were recorded by using a steady state flow concentrations ranging from 0.1 to 0.0005 mg mL
1. Then 60 mL of
ramp in the 0.01e1000 s
1 range of shear rate. The shear rate was ANS (8.0 mM in 0.01 M phosphate buffer, pH 7.0) was added to 3 mL
recorded point by point with consecutive 3 min steps of constant of sample. Relative fluorescence intensity (RFI) was measured with
shear rate. This time allowed obtaining constant shear rates for an RF-5301 PC spectrofluorometer (Shimadzu Corp., Kyoto, Japan),
each point. The viscosity was then determined for each point and at wavelengths of 365 nm (excitation) and 484 nm (emission). The
the flow curve could be built. The values for n and k were obtained initial slope of RFI versus protein concentration (mg mL
1) (calcu-
from plots of log shear stress versus log shear rate, according to the lated by linear regression analysis) was used as an index of the
power law equation (Jambrak et al., 2009) log s ¼ log k þ n log g, protein hydrophobicity.
where s is the shear stress (Pa); g is the shear rate (s
1); n is the
flow behaviour index, and k is the consistency index (Pa sn). 2.8. Scanning electron microscopy (SEM)
Apparent viscosity (happ) was calculated at 0.05 s
1 from the rela-
tionship s ¼ happg using Newtonian law and the linear least square The morphology of the lyophilized SPI samples was observed
method for regression analysis. with a scanning electron microscope (JSMe6390 LV, Japan) at an
accelerating voltage of 5 kV. Before using the scanning electron
2.5. Determination of protein solubility microscope, the samples were coated with gold/palladium in an
argon atmosphere using a Balzers evaporator (model SCD 050,
The protein solubility of samples in different solvents was Baltec Lichtenstein, Austria) (Favaro-Trindade, Santana, Monterrey-
determined by modification of the methods described by Shimada Quintero, Trindade, & Netto, 2010).
and Cheftel (1988), Yin, Tang, Wen, and Yang (2007) and Manoi and
Rizvi (2009), as follows. Lyophilized samples (2 mg mL
1) were
2.9. Sodium dodecyl sulphate-polyacrylamide gel electrophoresis
dispersed in various solvents as follows: DW, deionized water at pH
(SDS-PAGE)
8.0 (adjusted with NaOH); Buffer B, Triseglycine buffer (0.086 M
Tris, 0.09 M glycine, and 4 mM Na2EDTA, pH 8.0); Buffer BSU, Buffer
Sodium dodecyl sulphate-polyacrylamide gel electrophoresis
B containing 0.5% sodium dodecyl sulphate (SDS) and 6 M urea;
was performed on a discontinuous buffered system with mercap-
Buffer BSUM, Buffer BSU plus 1% (v/v) b-mercaptoethanol (2-ME).
toethanol according to the method of Laemmli (Laemmli, 1970;
The mixtures were incubated for 24 h at 25  C in a shaking water
Zhao, Liu, Zhao, Ren, & Yang, 2011), using 10% separating gel and
bath. The resultant solutions were centrifuged at 20,000g for
4% stacking gel. Lyophilized SPI samples (2 mg mL
1 in buffer
20 min at 25  C, and the protein concentration in the supernatants
containing 0.0625 M TriseHCl, 10% glycerin, 2% SDS, and 5% 2-
was determined by the Lowry method using bovine serum albumin
mercaptoethanol, 0.0025% bromophenol blue) were incubated for
(BSA) as the standard. Absorbance at 750 nm was measured using
1 h at room temperature, then heated at 95  C for 5 min and
a T6 spectrophotometer (Beijing Purkinje General Instrument Co.
centrifuged at 5000 g for 10 min using a TGL-16G 144centrifuge
Ltd, Beijing, China). All determinations were conducted in dupli-
(Anting Scientific Instrument Co. Ltd., Shanghai, China). Aliquots
cate. Protein solubility (%) was calculated as 100  protein content
(15 mL) of the prepared samples were loaded onto the gels.
of the supernatant/total protein content.

2.6. Determination of SH content 2.10. Circular dichroism (CD) spectra measurement

The sulfhydryl (SH) content of the soluble fraction of lyophilized CD spectra were scanned at the far-UV range (260e180 nm)
samples was determined using Ellman’s reagent DTNB according to with a CD spectropolarimeter (Jasco 810, Jasco Corp., Tokyo,
the method described by Shimada and Cheftel (1988, 1989) and Japan) in a 0.1 cm quartz CD cuvette (Hellma, Muellheim, Baden,
Lagrain, Brijs, and Delcour (2008), with some modification. Germany) at 25  C. Freeze-dried SPI was dissolved in 0.01 M sodium
Lyophilized SPI samples were solubilized in pH 8.0 buffer (0.086 M phosphate buffer (pH 7.0) and centrifuged at 8000g at 25  C
Trise0.09 M glycine 4 mM Na2EDTA) at a protein concentration of temperature for 15 min to remove any insoluble residues. The
0.2% (0.1 g of protein/50 mL of buffer). The mixtures were incubated concentration of protein for CD analysis was 0.1 mg mL
1. The
for 24 h at 25  C in a shaking water bath, and then centrifuged at values of scan rate, response, and bandwidth were 100 nm min
1,
20,000g at 4  C for 15 min. The supernatant fractions were analyzed 0.25 s and 1.0 nm, respectively. Three scans were averaged to obtain
for SH group content. To a 3-mL aliquot of the supernatant was one spectrum. The secondary structure of SPI in this study was
added 0.03 mL of Ellman’s reagent solution (4 mg DTNB/mL buffer). estimated using the Yang-Us, jwr software provided by Jasco Corp.
After the solution was rapidly mixed and allowed to stand at 20  C Four secondary structures were calculated: a-helix, b-sheet, b-turn,
for 15 min, the absorbance was read at 412 nm. The buffer was used random coil (Kim, Kim, Yang, & Kwon, 2004).
instead of protein solutions as a reagent blank. A protein blank was
used in which 0.03 mL of the buffer replaced Ellman’s reagent 3. Results and discussion
solution. A molar extinction coefficient of 1.36  104 M
1 cm
1 was
used for calculating micromoles of SH/gram of protein. 3.1. Rheological measurement

2.7. Surface hydrophobicity (H0) measurements 3.1.1. Stress sweep test analysis
Fig. 1 presents the storage modulusestrain profiles of SPI
H0 was determined using 1-anilino-8-naphthalene-sulfonate dispersion (12.5% w/v) after ultrasonic treatment at the different
(ANS) as a fluorescence probe according to the method of Kato and combinations of power (200, 400, 600 W) and time (15 and
Nakai (1980) and others (Gulseren et al., 2007; Kato & Nakai, 1980), 30 min). For all ultrasonic treatments investigated, G0 remained
in the absence of SDS. Lyophilized SPI samples (1 mg mL
1 in almost constant as strain increased and then suddenly decreased.
0.01 M phosphate buffer at pH 7.0) were centrifuged at 20,000g at The strain amplitude at which G0 just began to decrease by 5% from
4  C for 15 min. After determining the protein concentration in the its maximum value was determined and taken as a measure of the
supernatants according to the method of Lowry, each supernatant limit of linearity critical strain of the gel (Maltais, Remondetto, &
650 H. Hu et al. / Food Hydrocolloids 30 (2013) 647e655

Fig. 1. Typical stress sweep tests for 12.5% soy protein dispersion of different ultrasonic
treatment. Symbols A (,), B (B), C (6), D (7), E (>), F (-) and G (C), represent the
following sample treatments: no ultrasound (A); 200 W, 15 min (B); 200 W, 30 min (C);
400 W, 15 min (D); 400 W, 30 min (E); 600 W, 15 min (F); 600 W, 30 min (G).

Subirade, 2008). Thus, in our following study, the strain % was

selected as 0.05%.

3.1.2. Frequency sweep analyses

Fig. 2 shows the influence of the different ultrasonic treatments
on the variation of the storage modulus (G0 ), loss modulus (G00 ), and
phase angle (d) of the SPI dispersions (12.5% w/v). It could be
observed that both G0 and G00 of all of the samples increased with
increasing angular frequency. Similar frequency dependencies for
the G0 and G00 moduli have been observed for various types of
protein dispersions (Madadlou et al., 2010; Sittikijyothin et al.,
2010).
As can be seen from Fig. 2, ultrasonic treatment decreased both
G0 and G00 of SPI dispersions, similar to what has been reported for
wheat gluten (Zhang, Claver, Zhu, & Zhou, 2011). One of the possible
reasons is that ultrasonic treatment can reduce the particle size of
SPI dispersions (Jambrak et al., 2009), and particle size distribution
had its manifestation on the flow properties (Arzeni et al., 2011).
Furthermore, the d value (tan d ¼ G00 /G0 ) of treated SPI was much
higher than untreated SPI. These results indicated that at a protein
concentration of 12.5% (w/v), untreated SPI formed a more elastic
dispersion (solid nature), while the ultrasonic-treated SPI samples
formed more viscous dispersions (fluid character). The above
results suggest that ultrasonic treatment decreased the intermo-
Fig. 2. Influence of different ultrasonic treatments on the variation of the storage
lecular interactions between SPI molecules, possibly by weakening modulus (G0 ), loss modulus (G00 ), and phase angle (d) of SPI dispersions. Symbols A (,),
of non-covalent bonding such as hydrogen bond between protein B (B), C (6), D (7), E (>), F (-) and G (C) represent samples as identified in the
molecules (Section 3.2). caption to Fig. 1.

3.1.3. Flow curves of determination) values. All SPI dispersions exhibited a pseudo-
Fig. 3 shows the influence of different ultrasonic treatments on plastic behaviour (n < 1) over the range of 10e1000 s
1. Ultrasound
the variation of flow curves of SPI dispersions. Untreated SPI solu- treatment significantly decreased the consistency coefficients and
tion and SPI sonicated at 200 W for 15 min (sample B) exhibited increased the flow behaviour index of SPI dispersions. Similar
pseudoplastic behaviour over the entire shear rate range (0.01e results were reported by Karki et al. (2009). During ultrasound
1000 s
1). A pseudoplastic material is one in which viscosity treatment, there is rapid molecule movement due to cavitation and
decreases with increasing rate of shear (also termed shear thin- microstreaming and unfolding of protein chains, leading to the
ning). In contrast, the other sonicated dispersions were shear changes in the flow behaviour of SPI. In contrast, Jambrak et al.
thinning with a Newtonian region in the lower shear rate range. (2009) reported the opposite results, that ultrasound treatment
Table 1 shows the rheological properties of different ultrasonic- could increase the consistency coefficients and decrease the flow
treated SPI dispersions in higher shear rate range (10e1000 s
1). behaviour index. The difference might be that Jambrak et al. (2009)
The fitted curves for the SPI dispersions gave good agreement did not control the temperature during the ultrasound treatment,
with the experimental data, as illustrated by the high R2 (coefficient which could lead to increased temperature. In contrast, we carried
 H. Hu et al. / Food Hydrocolloids 30 (2013) 647e655 651

hydrophobic interactions and Buffer BSUM (BSU plus 1% (v/v) 2-

ME) can disrupt disulfide bonds (Yin et al., 2007).
Table 2 shows the solubility of treated and untreated SPI in the
different extraction media. It was observed that all ultrasonic
treatments (200e600 W, 15 and 30 min) could significantly
increase the solubility of SPI in DW (deionized water at pH 8.0) and
in Buffer B (Triseglycine buffer (0.086 M Tris, 0.09 M glycine, and
4 mM Na2EDTA, pH 8.0)) (P < 0.05). The protein solubility of SPI
increased significantly with increasing ultrasonic intensity and
ultrasonic time, from 9.98  0.41% in DW and 16.90  0.08% in
Buffer B for untreated SPI, to 54.78  0.73% in DW and 49.18  1.31%
in Buffer B solubility for SPI treated with 600 W for 30 min (sample
G). The increased solubility was not due to hydrolysis of peptide
bonds of the SPI protein molecules, since the SDS-PAGE profile of all
the samples were similar (see Section 3.6). Therefore, this increase
in solubility may be due to the conformational change during
ultrasonic treatment and formation of soluble protein aggregates
(Tang et al., 2009) from insoluble protein aggregates. Tang et al.
Fig. 3. Influence of different ultrasonic treatments on the variation of flow curves of
(2009) pointed out that ultrasonic treatment will facilitate the
SPI dispersions. Symbols A (,), B (B), C (6), D (7), E (>), F (-) and G (C) represent formation of soluble protein aggregates from insoluble precipitates
samples as identified in the caption to Fig. 1. in commercial SPI. Furthermore, ultrasonic treatment can reduce
the particle size of SPI, which increases proteinewater interactions,
resulting in increased protein solubility (Arzeni et al., 2011). Various
out our experiment with an ice-water bath. It was also observed
researchers (Chen et al., 2011a; Jambrak et al., 2009; Karki et al.,
that for pair of samples treated using the same ultrasound intensity
2009; Tang et al., 2009) had found ultrasound treatment can
but treated for different times (namely, samples B (200 W 15 min)
improve the protein solubility of SPI. In the case of ultrasonic-
and C (200 W 30 min): 75e83 W cm
2; samples D (400 W 15 min)
treated SPI (sample BeG), the solubility in Buffer B was slightly but
and E (400 W 30 min): 105e110 W cm
2; samples F (600 W 15 min)
significantly (P < 0.05) lower than that in DW. Similar phenomena
and E (600 W 30 min): 131e138 W cm
2), the longer ultrasound
had been reported in SPI films (Yin et al., 2007), but the reason for
treatment time could result in higher flow behaviour index and
the reduction of protein solubility from DW to Buffer B is not clear.
lower consistency coefficient, which could be due to an increase in
Solubility of all the SPI samples was increased by the presence of
the denaturation of SPI with the increased treatment time.
denaturing agents in Buffer BSU (Table 2). Although no significant
differences (P > 0.05) were observed between any of the SPI
3.2. Protein solubility
samples in terms of their protein solubility in Buffer BSU, the
difference of protein solubility in Buffer BSU compared to Buffer B
Solubility is the most practical measure of protein denaturation
was different for samples under different ultrasonic treatment. In
and aggregation, and hence a good index of protein functionality.
particular, it was found that the difference between solubility in
Urea is an agent known to disrupt non-covalent interactions, such
Buffer B and Buffer BSU was smaller after ultrasonic treatment,
as hydrogen bonds and hydrophobic interactions, while SDS is
suggesting that ultrasonic treatment could disrupt some of the
expected to associate with hydrophobic amino acids in the protein
non-covalent interactions of SPI, analogous to the disruption ach-
and thus interfere with hydrophobic proteineprotein interactions
ieved under denaturing conditions. This may be attributed to
(Shimada & Cheftel,1989). 2-ME is a strong reducing agent and
cavitation phenomenon which promotes a turbulent flow, high
generally can cleave disulfide (SeS) bonds. Thus, the comparison of
pressure and shear force. These physical factors might disrupt the
protein solubility in different extraction media or buffers with or
hydrogen bonds and hydrophobic interactions responsible for
without denaturing and/or reducing agents can provide informa-
intermolecular association of proteins (Maity, Rasale, & Das, 2012).
tion about the intermolecular bonds involved in protein cross-
Tian, Wan, Wang, and Kang (2004) inferred that the large interfacial
linking (Manoi & Rizvi, 2009). For example, Buffer B (Triseglycine
area between water and air produced by ultrasound cavitation
buffer (0.086 M Tris, 0.09 M glycine, and 4 mM Na2EDTA, pH 8.0))
disturbs the condition around the trypsin molecules, such as
typically disrupts the electrostatic interactions (Mauri & Añón,
hydrogen bond and hydrophobic interaction. Similarly, Zhang et al.
2006; Yin et al., 2007), while Buffer BSU (Buffer B containing 0.5%
(2011) also pointed out that ultrasonic treatment might weaken the
sodium dodecyl sulphate (SDS) and 6 M urea) also disrupts the
other non-covalent interactions such as hydrogen bonds and
Table 2
Protein solubility of untreated and ultrasonication-treated SPI in different solvents.a
Table 1 Each value represents the mean  standard deviation (n ¼ 3).
Rheological properties of different ultrasonication-treated SPI solutions in higher
shear rate range (10e1000 s
1). DW Buffer B Buffer BSU Buffer BSUM
A (untreated SPI) 10.0  0.4a 16.9  0.1a 60.4  3.2a 85.3  0.0b
Treatment Flow Consistency Coefficients B (200 W 15 min) 40.3  0.5b 37.4  0.7b 59.9  1.0a 81.8  3.3 ab
behaviour coefficient of determination C (200 W 30 min) 46.4  0.2c 42.9  0.0c 60.2  2.9a 84.4  1.2b
index (n) (k) Pa s (R2) D (400 W 15 min) 49.5  1.2d 44.3  1.2c 62.8  3.7a 74.3  3.3a
A (untreated SPI) 0.1975 52.5886 0.9609 E (400 W 30 min) 53.6  1.4ef 47.2  1.3d 58.7  1.7a 80.7  6.5 ab
B (200 W 15 min) 0.5691 2.1793 0.9999 F (600 W 15 min) 51.9  2.2de 47.0  2.0d 59.0  0.7a 79.8  0.4 ab
C (200 W 30 min) 0.6230 0.9329 0.9990 G (600 W 30 min) 54.8  0.7f 49.2  1.3d 61.6  1.0a 77.9  4.2 ab
D (400 W 15 min) 0.2714 17.3293 0.9915 a
DW, deionized water at pH 8.0; Buffer B, Triseglycine buffer (0.086 M Tris,
E (400 W 30 min) 0.5130 1.8078 0.9992
0.09 M glycine, and 4 mM Na2EDTA, pH 8.0); Buffer BSU, Buffer B containing 0.5%
F (600 W 15 min) 0.3881 4.6145 0.9924
sodium dodecyl sulphate (SDS) and 6 M urea; Buffer BSUM, Buffer BSU containing
G (600 W 30 min) 0.5463 1.3047 0.9997
1% 2-ME.
652 H. Hu et al. / Food Hydrocolloids 30 (2013) 647e655

non-covalent bonding between wheat gluten protein molecules. alter the thiol content of whey protein concentrate. Gulseren et al.
From the difference of SPI solubility in Buffer BSU compared to (2007) hypothesized that cavitation-generated hydrogen peroxide
Buffer B, we could infer that for samples B (200 W 15 min), C might oxidize susceptible free SH groups, resulting in the decrease
(200 W 30 min) and D (400 W 15 min) the non-covalent interac- of free SH group content of BSA, while Chandrapala et al. (2010)
tions (hydrogen bonds and hydrophobic interactions) of SPI were pointed out that the intra-molecular location of the free thiol
decreased with ultrasonic time and intensity. However, longer groups in b-lactoglobulin and a-lactalbumin might make them less
ultrasonic time (sample E: 400 W 30 min) and higher intensity susceptible to degradation by ultrasound. Besides, the differences
(samples F and G: 600 W 15 and 30 min) did not result in any could also be related to the complexity of WPC, which contained
further decrease in the hydrogen bonds and hydrophobic interac- a mixture of proteins rather than the pure BSA used by Gulseren
tions or additional improvement in solubility of the SPI samples. As et al. (2007). In this study, the increase of free SH content might
expected, all SPI (ultrasonic-treated and untreated) samples had infer that ultrasonic treatment could break disulfide bonds (i.e.
the highest protein solubility in Buffer BUSM (Buffer BUS plus 1% 2- cause reduction of SS to form SH groups). There are at least 20
ME), similar to the result of Shimada & Cheftel (1989). However, the disulfide groups in glycinin (11S protein) (Kinsella, 1979), which
BSUM data are a little puzzling, since some part of the SPI that was may be altered by ultrasonic treatment. Moreover, our recent
soluble in the untreated sample A became no longer soluble in the researches (Hu et al., 2012) found that ultrasonic treatment
sonicated sample D even in the presence of denaturing as well as reduced particle size of SPI in the dispersions, which might be
reducing agents. Perhaps this is an indication that there were some attributed to that intermolecular disulfide bonds were broken by
other types of covalent bonds that were formed after sonication. ultrasonic treatment. Interestingly, Arzeni et al. (2011) found that
This would be consistent with the larger aggregates seen in the SEM the free SH groups of SPI did not change significantly after ultra-
data (Section 3.5). sonic pretreatment. The difference of our results and theirs might
be due to the different temperature of the final ultrasonic SPI. They
3.3. Determination of free SH groups used a glycerin-jacketed circulating constant temperature cooling
bath to keep the temperatures at the end of sonication to below
Fig. 4 shows that free sulfhydryl content of soluble SPI was 49  C, while we used an ice-water bath to maintain the samples’
increased significantly after ultrasonic treatment (P < 0.05). temperature.
Besides, the SH groups content increased with increasing ultrasonic
time and intensity, from 9.13  0.44 mmol g
1 soluble protein 3.4. Surface hydrophobicity (H0)
(untreated SPI) to 18.08  0.39 mmol g
1 soluble protein (G 600 W
30 min). The increase of sulfhydryl content might be attributed to Protein surface hydrophobicity is an index of the number of
the exposure of sulfhydryl groups to the surface of SPI molecules. hydrophobic groups on the surface of a protein molecule in contact
Jambrak et al. (2009) and Arzeni et al. (2011) reported that ultra- with the polar aqueous environment and is closely related to its
sonic treatment can reduce the particle size of SPI. Thus it is functional properties (Chandrapala et al., 2010; Chen et al., 2011a).
possible that the buried sulfhydryl groups of SPI are exposed during Fig. 5 shows that ultrasonic treatment could significantly increase
the process of reducing SPI size, because the turbulent flow, high the surface hydrophobicity of soluble SPI (P < 0.05). This finding
pressure and shear force of cavitation phenomenon. Fernandez- was consistent with previous studies, which showed that ultra-
Diaz, Barsotti, Dumay, and Cheftel (2000) pointed out that elec- sound treatment could cause the increase of surface hydropho-
tric pulse processing resulted partial unfolding of the ovalbumin bicity for BSA (Gulseren et al., 2007) and SPI (Chen et al., 2011a).
protein, thus exposing sulfhydryl groups to the surface. The ultra- Comparing the surface hydrophobicity of SPI sonicated under
sonic treatment might have similar effect because ultrasonic can different conditions, it was observed that H0 was increased with
also partial unfold SPI protein. However, Gulseren et al. (2007) ultrasonic time (from 15 to 30 min) at the same ultrasonic intensity
found that the amount of free sulfhydryl groups in BSA decreased (200, 400 or 600 W, respectively) and was increased with ultrasonic
with increasing sonication time, while Chandrapala, Zisu, Palmer, intensity (from 200 to 600 W) at the same ultrasonic time (15 or
Kentish, and Ashokkumar (2010) reported that sonication did not 30 min). The above results indicate that before ultrasonic

Fig. 4. Content of free SH groups in untreated (A) and ultrasonication-treated (BeG) Fig. 5. Surface hydrophobicity of untreated (A) and ultrasonication-treated (BeG) SPI.
SPI. The sample codes are described in the caption to Fig. 1. Each value represents the The sample codes are described in the caption to Fig. 1. Each value represents the mean
mean and standard deviation (n ¼ 3). and range (n ¼ 2).
 H. Hu et al. / Food Hydrocolloids 30 (2013) 647e655 653

treatment, the hydrophobic regions of native SPI were located a magnification factor of 150-fold. It was observed that the samples
buried within the interior of the molecules, while the cavitation BeG obtained after ultrasonic treatments and freeze drying had
phenomenon induced by ultrasonic treatment could expose some larger and more heterogenous structures than sample A (untreated
of the hydrophobic regions of SPI to the surface. SPI). In addition, samples E (400 W 30 min) and G (600 W 30 min)
In this study, we found that the greater protein surface hydro- were larger than D (600 W 15 min) and F (600 W 15 min)
phobicity the greater the SPI solubility. Similarly, Wagner, Sorgentini, respectively suggesting that longer ultrasonic time could result in
and Anon (2000) studied the relationship between solubility and larger structures. These results might be due to the changes in
surface hydrophobicity and found that the greater surface hydro- ultrasonic treatment leading to unfolding of the SPI molecules, and
phobicity, the greater the solubility. In contrast, Hayakawa and Nakai increased exposure of free SH groups (Section 3.3) and hydro-
(1985) found that ANS hydrophobicity was inversely related to the phobic groups (Section 3.4) at the surface of the molecules, which
solubility of soy protein. These conflicting observations may be could interact with each other and form larger aggregates during
attributed to differences in the extent of intermolecular interactions freeze drying.
between the hydrophobic groups on the surface of the protein Usually, smaller aggregates in dispersions may result in a higher
molecules. Involvement of surface exposed hydrophobic groups in solubility (Tang et al., 2009). Our recent research (Hu et al., in press)
extensive intermolecular interactions would be expected to result in found that even though the aggregates of ultrasonic-treated
lower solubility. Conversely, the higher solubility of sonication samples in dry state are larger, the aggregates in dispersions are
treated SPI with increased surface hydrophobicity may be an indi- smaller. Besides, the particle size and particle distribution of
cation of decreased intermolecular interactions and smaller particle ultrasonic-treated SPI dispersions were smaller and different from
size of SPI dispersions after ultrasonication. the untreated samples. These findings were consistent with the
improvement of SPI protein solubility (section 3.2) after ultrasonic
3.5. Scanning electron microscopy (SEM) treatments.

In order to understand the effect of different ultrasonic treat- 3.6. SDS-PAGE

ments on SPI, the microstructure of lyophilized SPI was observed
by SEM. Fig. 6 shows a set of SEM images of different SPI at As shown in Fig. 7, untreated SPI (A) presented a typical SDS-
PAGE profile of individual subunits of glycinin (11S) and b-con-
glycinin (7S). Comparing ultrasonic-treated and untreated SPI,
ultrasonic treatment did not induce major changes in the protein
electrophoretic patterns, suggesting that ultrasonic treatment did
not modify the protein profiles of SPI regardless of the sonication
conditions in our experiment. Similar results were reported by
Chen et al. (2011a) who found that 400 W ultrasonic-treated SPI
and SPI displayed similar peptide profiles. Karki et al. (2009)
also reported that sonication did not modify the protein profile
of SPI.

Fig. 7. SDS-PAGE electrophoretic profiles of untreated (A) and ultrasonication-treated

Fig. 6. Microstructure of untreated (A) and ultrasonication-treated (BeG) lyophilized (BeG) SPI. The sample codes are described in the caption to Fig. 1. The bands corre-
SPI determined by scanning electron microscopy. The sample codes (AeG) are sponding to subunits of the 7S and 11S soy globulins are identified on the right side of
described in the caption to Fig. 1. the electrophoretogram.
654 H. Hu et al. / Food Hydrocolloids 30 (2013) 647e655

Table 3 Protein solubility was increased by ultrasonic treatment. The non-

Secondary structural content of native and ultrasonic-treated SPI estimated from covalent interactions decreased after ultrasonic treatments. The
circular dichroism spectra in the far-UV region (180e260 nm) at 25  C.
free sulfhydryl content was increased after ultrasonic treatment,
a-Helix (%) b-Sheet (%) b-Turn (%) Random coil (%) and protein surface hydrophobicity (H0) was also increased with
A (untreated SPI) 7.6 60.0 0.0 32.3 increasing ultrasonic time and intensity. No significant change was
B (200 W 15 min) 5.5 66.2 3.6 24.7 observed in the protein electrophoretic patterns (SDS-PAGE) but
C (200 W 30 min) 6.7 63.5 0.0 29.8
changes in the secondary structure of SPI by ultrasonication were
D(400 W 15 min) 7.7 63.2 3.1 26.0
E (400 W 30 min) 11.7 49.0 4.7 34.6 indicated by CD analysis. The effects of ultrasonic treatment on SPI
F (600 W 15 min) 8.8 58.2 2.1 30.9 were dependent on the intensity and time of ultrasonication. In
G (600 W 30 min) 14.0 38.2 8.8 38.9 conclusion, under the conditions investigated in this study, ultra-
sonic treatment resulted in partial unfolding and reduction of
intermolecular interactions as demonstrated by increases in free
3.7. Circular dichroism (CD) spectra sulfhydryl groups and surface hydrophobicity, leading to improved
solubility and fluid character of SPI solutions. Larger aggregates in
The secondary structure of SPI was determined using CD spec- the dry state were formed after freeze drying of ultrasonication-
troscopy for native and ultrasonic-treated dispersions, and the treated SPI. Controlling the ultrasonic treatment conditions could
proportions of a-helix, b-sheet, b-turn and random coil were derived result in SPI with different structural properties, and further studies
from Yang’s equation (Kim et al., 2004) are shown in Table 3. The should be conducted to investigate the impact of these changes on
secondary structure of untreated SPI was changed by ultrasonic gelation and other functional properties.
treatment, with greater changes at higher power and time treatments
(E 400 W 30 min, F 600 W 15 min and G 600 W 30 min compared to B Acknowledgements
200 W 15 min, C 200 W 30 min and D 400 W 15 min). Moreover,
compared with the untreated SPI, samples B (200 W 15 min) and C The authors would like to thank Mr. Wang Kexing for financial
(200 W 30 min) showed a decrease in a-helix and random coil, and an support and Dr. Jambrak for technical support.
increase in b-sheet. In contrast, samples E (400 W 30 min) and G
(600 W 30 min) showed an increase in a-helix and random coil, and
References
a decrease in b-sheet. The above results suggested that, the a-helix
and random coil structures of SPI were decreased after lower power Arzeni, C., Martínez, K., Zema, P., Arias, A., Pérez, O., & Pilosof, A. (2011). Compar-
ultrasonic treatment (200 W) and increased after higher power ative study of high intensity ultrasound effects on food proteins functionality.
treatment (600 W). Since the secondary structure of protein depends Journal of Food Engineering, 108(3), 463e472.
Chandrapala, J., Zisu, B., Palmer, M., Kentish, S., & Ashokkumar, M. (2010). Effects of
on both the local sequence of amino acids and the interactions ultrasound on the thermal and structural characteristics of proteins in recon-
between different parts of molecule, the above results may indicate stituted whey protein concentrate. Ultrasonics Sonochemistry, 18(5), 4198e4207.
that sonication can disrupt these interactions, leading to secondary Chen, L., Chen, J., Ren, J., & Zhao, M. (2011a). Effects of ultrasound pretreatment on the
enzymatic hydrolysis of soy protein isolates and on the emulsifying properties of
structure changes. Similarly, Chandrapala et al. (2010) found that hydrolysates. Journal of Agricultural and Food Chemistry, 59(6), 2600e2609.
ultrasonic treatment (20 kHz, 450 W) resulted in an increase in the a- Chen, L., Chen, J., Ren, J., & Zhao, M. (2011b). Modifications of soy protein isolates
helix component and a decrease in the b-sheet component of whey using combined extrusion pre-treatment and controlled enzymatic hydrolysis
for improved emulsifying properties. Food Hydrocolloids, 25(5), 887e897.
protein concentrate, while Stathopulos et al. (2004) pointed out that Favaro-Trindade, C., Santana, A., Monterrey-Quintero, E., Trindade, M., & Netto, F.
sonication-induced (20 kHz, 30 W) proteins had an increased b- (2010). The use of spray drying technology to reduce bitter taste of casein
structure with a concomitant decrease in a-helix structure in the hydrolysate. Food Hydrocolloids, 24(4), 336e340.
Fernandez-Diaz, M. D., Barsotti, L., Dumay, E., & Cheftel, J. C. (2000). Effects of pulsed
aggregates. The contradictory results might be due to different electric fields on ovalbumin solutions and dialyzed egg white. Journal of Agri-
protein fractions being analyzed. Stathopulos et al. (2004) investi- cultural and Food Chemistry, 48(6), 2332e2339.
gated samples prepared by centrifuging sonicated protein solutions Gulseren, I., Guzey, D., Bruce, B. D., & Weiss, J. (2007). Structural and functional
changes in ultrasonicated bovine serum albumin solutions. Ultrasonics Sono-
at 16,000g for 10 min and resuspending the resulting pellets in water.
chemistry, 14(2), 173e183.
However, Chandrapala et al. (2010) and we investigated the whole Guzey, D., Gulseren, I., Bruce, B., & Weiss, J. (2006). Interfacial properties and
protein solutions. Furthermore, Gulseren et al. (2007) found a small structural conformation of thermosonicated bovine serum albumin. Food
decrease in random coil content of bovine serum albumin (BSA) after Hydrocolloids, 20(5), 669e677.
Hu, H., Fan, X., Zhou, Z., Xu, X. Y., Fan, G., Wang, L. F., et al. Acid-induced gelation
ultrasonic treatment (20 kHz, 450 W), which was similar to our behavior of high intensity ultrasonic soybean protein isolate. Ultrasonics Sono-
results. Thus, the CD results suggested firstly that ultrasonic treat- chemistry, in press.
ment can change the secondary structure of SPI, and secondly, that Jambrak, A. R., Lelas, V., Mason, T. J., Kresic, G., & Badanjak, M. (2009). Physical
properties of ultrasound treated soy proteins. Journal of Food Engineering, 93(4),
higher power (400 W and 600 W) combined with longer time 386e393.
(30 min) could increase the a-helix component and decrease the b- Jambrak, A. R., Mason, T. J., Lelas, V., Herceg, Z., & Herceg, I. L. (2008). Effect of
sheet and random coil component. ultrasound treatment on solubility and foaming properties of whey protein
suspensions. Journal of Food Engineering, 86(2), 281e287.
Karki, B., Lamsal, B. P., Grewell, D., Pometto, A. L., Van Leeuwen, J., Khanal, S. K., et al.
(2009). Functional properties of soy protein isolates produced from ultra-
4. Conclusion sonicated defatted soy flakes. Journal of the American Oil Chemists’ Society,
86(10), 1021e1028.
Kato, A., & Nakai, S. (1980). Hydrophobicity determined by a fluorescence probe
Sonication of reconstituted SPI solutions resulted in structural
method and its correlation with surface properties of proteins. Biochimica et
and physical changes of SPI. Untreated SPI dispersion formed Biophysica Acta (BBA) e Protein Structure, 624(1), 13e20.
a more elastic solution (solid nature), while the ultrasonic-treated Kim, K. S., Kim, S., Yang, H. J., & Kwon, D. Y. (2004). Changes of glycinin confor-
mation due to pH, heat and salt determined by differential scanning calorim-
SPI dispersion formed more viscous solution (fluid character). The
etry and circular dichroism. International Journal of Food Science & Technology,
flow behaviour index (n) was increased after ultrasonic treatments, 39(4), 385e393.
and the longer ultrasound time could result in higher flow behav- Kinsella, J. E. (1979). Functional properties of soy proteins. Journal of the American
iour index. Scanning electron microscopy of lyophilized SPI showed Oil Chemists’ Society, 56(3), 242e258.
Kresi
c, G., Lelas, V., Jambrak, A. R., Herceg, Z., & Brn ci
c, S. R. (2008). Influence of
that the microstructure of ultrasonic-treated SPI was changed, with novel food processing technologies on the rheological and thermophysical
larger aggregates for ultrasonic-treated SPI than non-treated SPI. properties of whey proteins. Journal of Food Engineering, 87(1), 64e73.
 H. Hu et al. / Food Hydrocolloids 30 (2013) 647e655 655

Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the Sittikijyothin, W., Sampaio, P., & Gonçalves, M. (2010). Microstructure and rheology
head of bacteriophage T4. Nature, 227(5259), 680e685. of b-lactoglobulinegalactomannan aqueous mixtures. Food Hydrocolloids, 24(8),
Lagrain, B., Brijs, K., & Delcour, J. A. (2008). Reaction kinetics of gliadineglutenin 726e734.
cross-linking in model systems and in bread making. Journal of Agricultural and Soria, A. C., & Villamiel, M. (2010). Effect of ultrasound on the technological prop-
Food Chemistry, 56(22), 10660e10666. erties and bioactivity of food: a review. Trends in Food Science & Technology,
Madadlou, A., Emam-Djomeh, Z., Mousavi, M. E., Mohamadifar, M., & Ehsani, M. 21(7), 323e331.
(2010). Acid-induced gelation behavior of sonicated casein solutions. Ultrasonics Stathopulos, P. B., Scholz, G. A., Hwang, Y. M., Rumfeldt, J. A. O., Lepock, J. R., &
Sonochemistry, 17(1), 153e158. Meiering, E. M. (2004). Sonication of proteins causes formation of aggregates
Madadlou, A., Mousavi, M. E., Emam-Djomeh, Z., Ehsani, M., & Sheehan, D. (2009). that resemble amyloid. Protein Science, 13(11), 3017e3027.
Comparison of pH-dependent sonodisruption of re-assembled casein micelles Tang, C. H., Wang, X. Y., Yang, X. Q., & Li, L. (2009). Formation of soluble aggregates
by 35 and 130 kHz ultrasounds. Journal of Food Engineering, 95(3), 505e509. from insoluble commercial soy protein isolate by means of ultrasonic treatment
Maity, I., Rasale, D. B., & Das, A. K. (2012). Sonication induced peptide-appended and their gelling properties. Journal of Food Engineering, 92(4), 432e437.
bolaamphiphile hydrogels for in situ generation and catalytic activity of Pt Tian, Z. M., Wan, M. X., Wang, S. P., & Kang, J. Q. (2004). Effects of ultrasound and
nanoparticles. Soft Matter, 19(8), 5301e5308. additives on the function and structure of trypsin. Ultrasonics Sonochemistry,
Maltais, A., Remondetto, G. E., & Subirade, M. (2008). Mechanisms involved in the 11(6), 399e404.
formation and structure of soya protein cold-set gels: a molecular and supra- Wagner, J. R., Sorgentini, D. A., & Anon, M. C. (2000). Relation between solubility and
molecular investigation. Food Hydrocolloids, 22(4), 550e559. surface hydrophobicity as an indicator of modifications during preparation
Manoi, K., & Rizvi, S. S. H. (2009). Physicochemical changes in whey protein processes of commercial and laboratory-prepared soy protein isolates. Journal
concentrate texturized by reactive supercritical fluid extrusion. Journal of Food of Agricultural and Food Chemistry, 48(8), 3159e3165.
Engineering, 95(4), 627e635. Wang, X. S., Tang, C. H., Li, B. S., Yang, X. Q., Li, L., & Ma, C. Y. (2008). Effects of high-
Mauri, A. N., & Añón, M. C. (2006). Effect of solution pH on solubility and some pressure treatment on some physicochemical and functional properties of soy
structural properties of soybean protein isolate films. Journal of the Science of protein isolates. Food Hydrocolloids, 22(4), 560e567.
Food and Agriculture, 86(7), 1064e1072. Yin, S. W., Tang, C. H., Wen, Q. B., & Yang, X. Q. (2007). Properties of cast films from
Mu, L., Zhao, M., Yang, B., Zhao, H., Cui, C., & Zhao, Q. (2010). Effect of ultrasonic hemp (Cannabis sativa L.) and soy protein isolates. A comparative study. Journal
treatment on the graft reaction between soy protein isolate and gum acacia and of Agricultural and Food Chemistry, 55(18), 7399e7404.
on the physicochemical properties of conjugates. Journal of Agricultural and Zhang, H., Claver, I. P., Zhu, K. X., & Zhou, H. (2011). The effect of ultrasound on the
Food Chemistry, 58(7), 4494e4499. functional properties of wheat gluten. Molecules, 16(5), 4231e4240.
Shimada, K., & Cheftel, J. C. (1988). Determination of sulfhydryl groups and disulfide Zhao, G., Liu, Y., Zhao, M., Ren, J., & Yang, B. (2011). Enzymatic hydrolysis and their
bonds in heat-induced gels of soy protein isolate. Journal of Agricultural and effects on conformational and functional properties of peanut protein isolate.
Food Chemistry, 36(1), 147e153. Food Chemistry, 127(4), 1438e1443.
Shimada, K., & Cheftel, J. C. (1989). Sulfhydryl group/disulfide bond interchange Zisu, B., Bhaskaracharya, R., Kentish, S., & Ashokkumar, M. (2010). Ultrasonic pro-
reactions during heat-induced gelation of whey protein isolate. Journal of cessing of dairy systems in large scale reactors. Ultrasonics Sonochemistry, 17(6),
Agricultural and Food Chemistry, 37(1), 161e168. 1075e1081.