Food Hydrocolloids 39 (2014) 301e318

Contents lists available at ScienceDirect

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

Review

Soy proteins: A review on composition, aggregation and

emulsication
K. Nishinari a, *, Y. Fang a, *, S. Guo b, G.O. Phillips a
a
Glyn O. Phillips Hydrocolloids Research Centre, School of Food and Pharmaceutical Engineering, Faculty of Light Industry, Hubei University of Technology,
Wuchang, Wuhan 430068, PR China
b
College of Food Science and Nutritional Engineering, China Agricultural University, Beijing 100083, PR China

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

Article history: Composition of soybean proteins is briey described. Gels and gelling processes of soybean proteins and
Received 3 November 2013 other functionalities such as colloidal properties and emulsifying properties are described. The effects of
Accepted 6 January 2014 temperature, pH, ionic strength, processing conditions such as high pressure, ultrasonic treatment,
utilisation of enzyme, chemical modication are also described since they have been found useful to
Keywords: improve the processing and nal product.
Soy
2014 Elsevier Ltd. All rights reserved.
Protein
Gel
Emulsion
Process

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302
2. Main components of soybean proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302
3. Functionality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303
3.1. Solubility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303
3.2. Heat-induced denaturation and gelation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304
3.2.1. Effect of pH on gelation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307
3.2.2. Effects of salts on gelation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308
3.2.3. Effect of protein concentration on the elastic modulus of soy gels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308
3.2.4. Transparency of soy gels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309
3.2.5. Effect of coagulants on gels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309
3.2.6. Transglutaminase induced gelation of soybean globulins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 310
3.2.7. Cold-set gels of soy protein . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 310
3.2.8. How to enhance the gelling ? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311
3.3. Emulsification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312
3.3.1. Relation between hydrophobicity and emulsifying . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312
3.3.2. Emulsions stabilized by soy proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313
3.3.3. How to enhance the emulsification? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314
3.3.4. Emulsion gels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315
4. Concluding remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .316
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316

* Corresponding authors. Tel./fax: 86 (0)27 88015996.

E-mail addresses: katsuyoshi.nishinari@gmail.com (K. Nishinari), y.fang@
glyndwr.ac.uk (Y. Fang).

0268-005X/$ e see front matter 2014 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.foodhyd.2014.01.013
302 K. Nishinari et al. / Food Hydrocolloids 39 (2014) 301e318
1. Introduction
Soybeans have been cultivated for more than 3000 years in China
and other Asian countries, such as Japan and Korea. Some trials to
cultivate soybeans have been known in France and England since the
18th century, but have not been developed further. Since 1930, USDA
developed the cultivation and now USA has the largest production in
the world: USA, 7  107 t; Brazil,5.8  107 t; Argentine,5.8  107 t;
China, 1.7  107 t; India, 1.0  107 t (Kitamura, 2010). Soybeans have
been an important protein source in Asian countries and have been
utilised in various forms such as tofu (soybean curd), miso (fer-
mented soybean paste), natto (fermented soybeans covered with
mucilagenous substance), aburage (fried sheet of tofu) etc
(Nishinari, 1988). Recipe books on more than 100 different tofu
dishes were published in the Edo era (18th century) in Japan. In
addition to these traditional foods, an increased amount of soybean
milk is now consumed in Japan and in China due to its expected
health benet. Fibrous texture was also introduced in tofu-like
foods, making it resemble meat-like foods. Chen, Yamaguchi, and
Ono (2009) recently shed new light upon the formation of yuba, a
lm-like soybean food made from heated soymilk that contains oil
bodies, particulate protein, soluble protein, and carbohydrate.
The advantages of soybean proteins are: 1) provides a good
balance in amino acid composition, since all the essential amino
acids are contained, 2) contains physiologically benecial compo-
nents which are shown to lower the cholesterol, and reduce the risk
of hyperlipidemia and cardiovascular diseases, 3) has excellent
processing ability such as gelling, emulsifying ability and water-
and oil- holding capacity.
Soybeans should be heated before use in order to 1) deactivate
physiological harmful substances, such as trypsin inhibitor, and
hemaglutinin, 2) induce the denaturation of soybean protein, 3)
soften the tissue of soybean, 4) remove or reduce the raw soybean
odor, 5) to sterilize (Watanabe, Ebine, & Ota, 1991).
In addition to protein and oil, physiologically benecial effects of
daizein, isoavone in soybeans have been attracting much atten-
tion (Kitamura, 2010). Soluble soybean polysaccharides extracted
from residue (okara) in tofu-curd production have been shown to
be a good emulsier and have been widely used in the food in-
dustry (Kitamura, 2010).
2. Main components of soybean proteins
Soybean contains approximately 40% protein and 20% oil on
an average dry matter base. By removing oil at lower temperatures,
soy protein isolate (SPI) is obtained, and is widely used in the
food industry. Whole aqueous extractable soybean proteins can be
separated into storage globulin and whey fractions by acidication
to pH 4.5e4.8. The acid precipitable fraction includes the major
soybean storage proteins, and which is the main material consid-
ered in the present paper. The remaining part consists of the minor
globulin g-conglycinin, and relatively large amounts of contami-
nating proteins, including whey proteins which make up 9e15.3%
of soybean protein (Smith, Rackis, Isnardi, Cartter, & Krober, 1966).
Whey proteins are composed of lipoxygenase (LOX, 102 kDa), b-
amylase (61.7 kDa), lectin (33 kDa), and Kunitz trypsin inhibitors
(KTI, 20 kDa) (Iwabuchi & Yamauchi, 1987; Koshiyama, Kikuchi, &
Fukushima, 1981; Rackis, Wolf, & Baker, 1986). The proportion
represented by these whey proteins in the acid precipitated glob-
ulins is unknown (Pearson, 1983; Sorgentini & Wagner, 1999).
SPI is a mixture of various proteins, and the main ingredients are
classied into four protein categories according to their sedimen-
tation coefcients 2S, 7S, 11S and 15S which sediment at different
gravitational forces when the solution is subjected to a centrifugal
eld. In the present review, (7S and b-conglycinin) and (11S and
glycinin) are used interchangeably, and the history of the name was
described in Peng, Quass, Dayton, & Allen, (1984). Among these four
proteins, 7S (b-conglycinin) and 11S (glycinin) represent more than
80%, and the ratio 7S/11S has been reported to be about 0.5e1.3
depending on varieties (Saio, Kamiya, & Watanabe, 1969).
7S globulin consists of three subunits a (ca 67 kDa), a0 (ca 71 kDa)
and b (ca 50 kDa). 11S globulin is a hexamer, and is made up of ve
different subunits, each of which consists of an acidic subunit A
(acidic pI) with a molecular mass about 35 kDa and a basic subunit B
(basic pI) of molecular mass about 20 kDa, linked by a disulde
bond. AB subunits are believed to associate into two hexagonal rings
forming a hollow cylinder by electrostatic and hydrogen bondings
as shown schematically in Fig. 1 (Badley et al., 1975; Peng et al.
1984). Glycinin (11S) was found to dissociate into 2S, 3S or 7S
forms in various pH and ionic strengths (see Fig. 2).
Amino acid compositions of b-conglycinin and glycinin have
been analysed, but crystallization was difcult and the three
dimensional structure is not well established (Utsumi, Matsumura,
& Mori, 1997) in spite of many efforts. The crystal structure of 7S
and 11S have been recently studied by X-ray diffraction (Adachi
et al., 2003; Maruyama et al., 2001), and the previously proposed
picture of 11S was reconrmed and rened. They proposed that the
movement of a mobile disordered region to the side of the trimer,
and the dissociation of the hexamer into trimers may be susceptible
to proteinases (Adachi et al. 2003). Native glycinin is known to have
a compact structure stabilized by disulde bonds and thus its
emulsifying and foaming ability is lower than that of b-conglycinin
which lacks disulde bonds.
Ren, Tang, Zhang, and Guo (2009b) analyzed the aggregation
mode of polypeptides in protein particles of soy milk by using ul-
tracentrifugation, gel ltration, and sodium dodecyl sulfate poly-
acrylamide gel electrophoresis (SDS-PAGE). They proposed the
interaction mechanism of polypeptides in heat-induced protein
particles of soy milk: The proteins in soy milk dissociated, rear-
ranged, and aggregated to form protein particles when heated. The
protein particles of >40 nm in diameter dissociated into protein
aggregates with various molecular masses, which were dissociated
into monomeric subunits of 7S and 11S protein after treatment by
the mixture of 6 M urea and 0.5% SDS. The aggregates were pri-
marily composed of the disulde-linked basic and acidic poly-
peptides of 11S, besides a very small amount of a and a subunits of
7S. These aggregates and a part of monomeric subunits of 7S and
11S, as structural units, interact with each other to form protein
Fig. 1. Schematic diagram of glycinin molecule consisting of acidic, A, and basic, B,
subunits. (Badley et al., 1975).
 K. Nishinari et al. / Food Hydrocolloids 39 (2014) 301e318
Fig. 2. SDSePAGE shows the protein fractions prepared from heated and defatted soy
our. Soymilk, as a standard, containing 2 ng of nitrogen was applied to a well. The
amount that corresponded to the yield of each fraction was applied. Lane 1, marker;
lane 2, soymilk; lane 3, 11S; lane 4, LP; lane 5, 7S; lane 6, soy whey; AS, acidic subunit;
BS, basic subunit (Samoto et al., 2007).
particles primarily via noncovalent interactions, especially hydro-
phobic interactions and hydrogen bonding. The disulde-linked
basic polypeptides of the aggregates should be located inside the
protein particles, whereas the acidic, a and a subunits should be
located outside them for their high hydrophilicity.
Recently, the classication of soybean globulin was contested,
and the existence of a lipophilic protein (LP) in addition to 7S and
11S was proposed (Samoto et al., 2007). The protein content of LP
Fig. 3. Diagrammatic depiction of thermal aggregation behavior of b-conglycinin and glycinin at pH 7.0. N, native state; U, unfolded state; Agg., aggregates. (Guo, Yang, et al., 2012;
Guo, Zhang, et al., 2012).
303
was reported to be 76% which is lower than the 87% of 7S and 93% of
11S. However, LP showed a higher content of lipid 11.7% than 0.8% of
7S and 3.3% of 11S. (Samoto et al., 2007) reported that the LP yield
decreased, and simultaneously the yield of residue increased
although the yield of 7S and 11S was not changed with increasing
temperature. They attributed this change to the acceleration of
aggregation of LP by heating because of the hydrophobic properties
of LP.
Wu, Murphy, Johnson, Fratzke, and Reuber (1999) reported a
pilot plant fractionation to produce kilogram quantities of 11S, 7S
and an intermediate mixture using ultraltration rather than acid
precipitation. They also reported the effects of reducing agents and
salts concentration on the fractionation, yield and purity of soybean
storage protein fractions (Deak, Murphy, & Johnson, 2006).
Glycinin was recently isolated from soybeans using a mono-
clonal antibody with a yield of 16.8% and a purity of 93.8% based on
immunoafnity chromatography, which were signicantly higher
than those produced using other traditional procedures (Hu, Liu,
et al., 2013).
3. Functionality
3.1. Solubility
During soymilk processing, the soy proteins are prone to
denature and tend to interact with each other when heated. Many
research groups have investigated the aggregation of soy proteins
using pure proteins. The effect of 7S on the thermal aggregation of
11S has been examined. The addition of isolated 7S has been re-
ported to prevent the thermal aggregation of both 11S and the
isolated basic subunits of 11S (Damodaran & Kinsella, 1982). The
preferential association of the basic subunits of 11S and the b
subunits of 7S through an electrostatic force when 11S and 7S are
heated together has been reported (Utsumi, Damodaran, & Kinsella,
1984). The N-linked carbohydrate moieties of the a and a0 subunits
of 7S could prevent the heat-induced associations of 7S (Maruyama
et al. 2002; Utsumi, Maruyama, Satoh, & Adachi, 2002).
Isolated glycinin aggregates when heated at 80  C (Hashizume &
Watanabe,1979). The ultracentrifugal pattern showed that heating of
304 K. Nishinari et al. / Food Hydrocolloids 39 (2014) 301e318

acid-precipitated soy protein at 80  C resulted in the disappearance

of glycinin and the concurrent appearance of soluble aggregates and
protein components of 2e4 S. (Damodaran & Kinsella 1982) found a
similar function for egg albumin and bovine serum. They attributed
the prevention of aggregation by b-conglycinin to the formation of a
soluble complex between the subunits of b-conglycinin and the basic
subunits of glycinin via electrostatic interaction.
Guo, Yang, et al. (2012) studied the thermal aggregation kinetics
of b-conglycinin and glycinin at pH 7.0 using size exclusion chro-
matography with low-angle light scattering (SEC-LALS) to reveal
the assembly process of the soy protein unfolding polypeptide
chains during heating and clarify the role of b-conglycinin in the
thermal aggregation of soy protein (Fig. 3). They also characterized
the structure of the aggregates by small-angle X-ray scattering
(SAXS) and dynamic light scattering (DLS), to know the confor-
mational difference between soluble and insoluble aggregates.
They found that size of both b-conglycinin and glycinin increased
with increasing heating temperature, and the size of the mixture Fig. 4. Solubility of normal soy protein isolate (C) and acid-soluble soy protein
glycinin/b-conglycinin decreased with increasing b-conglycinin (B) (Hongo (Kitamura, 2010), in All about Soy protein, 2010, p. 487, in Japanese).
content. Unlike the b-conglycinin soluble aggregates that possessed
limited size and less compact conformation, particles with a denser They have also shown that lipoxygenase (LOX), b-amylase, and
core and a less dense outer shell were found in the glycinin insol- lectin of WSP and the b subunit of 7S tend to form a particulate
uble aggregates. They showed that growth of the size of the fraction, while the Kunitz trypsin inhibitors (KTI), a, and a0 subunits
insoluble aggregates of glycinin was terminated by b-conglycinin, tend to form a soluble fraction. To clarify which proteins increase
which is in good agreement with (Damodaran & Kinsella, 1982). protein particles in heated WSP with increasing temperature from
Guo, Yang, et al. (2012) attributed this function of b-conglycinin 65  C to 75  C, the changes in soluble protein constituent in heated
to its interaction with the least soluble basic polypeptides of gly- WSP after being heated from 60 to 95  C were analyzed by SDS-
cinin which improves its solubility, and then is consistent with PAGE. The result is shown in Fig. 6.
previous studies (Kuipers et al., 2006).
The main storage proteins in soybeans are globulins, and are
3.2. Heat-induced denaturation and gelation
insoluble in water, but soluble in dilute solutions of neutral salt. Soy
globulins show the lowest solubility between pH 4 and pH 5.
Although the gelation processes of linear macromolecules such
Since the heating of soybeans makes the globulins more insoluble, it
as polysaccharides and a brous protein such as gelatin have been
is necessary to know the solubility to develop their utilisation.
extensively studied and are comparatively well understood, the
Generally, the solubility of soybean globulins has been determined by
gelation of globular proteins is poorly understood because of its
the analysis of the protein fraction remaining in the supernatant after
complexity (Clark, 1998; Djabourov, Nishinari, & Ross-Murphy,
the solution is subjected to centrifugal forces. The degree of solubility
2013; Doi, 1993). When the globular proteins are denatured, they
is usually represented by nitrogen solubility index (NSI) dened;
convert into various intermediate states, such as pre-molten
globule, molten globule, brils, a subject which is still a matter of
total nitrogen content in soluble fraction debate in spite of tremendous activities. Dobson (2003) proposed a
which is extracted from the sample general model to describe the denaturation and aggregation
NSI
total nitrogen content in the sample

Tsumura et al. (2005) and Jung, Murphy, and Johnson, (2005)

reported that partially enzyme-hydrolysed soy globulins showed
higher solubility at around pH 4.5 but had lower solubility at
pH < 3 in comparison with non-hydrolysed globulins.
Recently, acid-soluble soy protein was developed so that it can
be used in acidied drinks as shown in Fig. 4.
Tofu-curd was traditionally made by adding calcium to soymilk
at low pH. The protein solubility of soymilk as a function of added
calcium has been shown to decrease sharply at 6 mM and 8 mM
calcium, and the protein solubility of soymilk as a function of pH
has been shown to decrease at pH 6.1, and pH of soymilk was
shown to decrease by adding calcium (Fig. 5) (Guo, Ono, & Mikami,
1999; Ono, Katho, & Mothizuki, 1993; Yuan et al., 2013).
Although there have been many studies on the solubility of
soybean proteins using fractionated 7S and 11S, it is also necessary
to understand the interaction between different ingredients in
soybeans. Ren, Tang, Zhang, and Guo (2009a) studied the interac-
tion between whey soybean protein (WSP) and 7S during the for-
mation of protein particles at elevated temperatures. This indicated
that 7S and WSP interact with each other during heating and form Fig. 5. Changes in protein solubility and separable oating fraction ratio in soy milk
complex protein particles which have special surface properties by addition of calcium chloride: (,)protein solubility; (B) oating fraction ratio; (6)
that are different from individual 7S and WSP protein particles. pH (Guo et al., 1999).
 K. Nishinari et al. / Food Hydrocolloids 39 (2014) 301e318
Fig. 6. Changes in the constituents of soluble protein fractions in heated WSP
dispersions (5 mg/mL protein). All of the proteins were dispersed in the ultraltrate of
raw soymilk before heat treatment at 60, 65, 70, 75, 80, 85, 90, and 95  C for 10 min,
respectively (Ren et al., 2009a).
process of globular proteins. Here the hydrophobic groups buried in
the native state are exposed to the surface, and the surface hy-
drophobicity increases, and net charge decreases. The b-sheet for-
mation is promoted leading to the network formation (Chiti &
Dobson, 2009). Recently, the gelation mechanism of globular pro-
teins has attracted much attention because common aspects have
been found with amyloid formation related with Creutzfeld-Jacobs
disease or bovine spongiform encephalopathy (BSE). Many food
proteins such as b-lactoglogulin, lysozyme and ovalbumin: all form
amyloid brils after denaturation, and these amyloids form a gel
above a certain critical concentration (Adamcik & Mezzenga, 2012;
van der Linden & Sagis, 2001).
Akkermans et al. (2007) reported amyloid brils from 11S and
SPI, and Tang and Wang (2010) reported those from 7S, 11S and
their mixtures. Amyloid brils are ordered aggregates of peptides
or proteins that are brillar in structure and contribute to the
complications of many diseases (e.g., type 2 diabetes mellitus,
Alzheimers disease, and primary systemic amyloidosis) (Nilsson,
2004). Nilsson categorized techniques to study amyloid bril for-
mation in vitro, and proposed three criteria that dene a protein
aggregate as an amyloid bril: green birefringence upon staining
with Congo Red, brillar morphology, and b-sheet secondary
structure.
While chaotropes such as urea and guanidine hydrochloride are
denaturing agents, for the soybean protein products in the food
industry, the denaturation is mainly induced by heating, freezing,
acid, high pressure and enzymes. When the globular protein so-
lutions are heated, the unfolding of molecular chains occurs and
hydrophobic groups buried in the internal part are exposed to the
surface. This denaturation phenomenon has been detected by dif-
ferential scanning calorimetry (DSC), circular dichroism (CD) and
Fourier Transform Infra-Red (FTIR) spectroscopy, transmission
electron microscopy (TEM) and atomic force microscopy (AFM)
(Clark, 1998). Bikbov et al. (1981) reported endothermic peaks at
68.5 and 84.5  C using a differential adiabatic microcalorimeter
with a sufcient slow heating rate (2  C/min) for a 0.5% solution of
soybean globulin solutions, which were attributed to the denatur-
ation of 7S and 11S, respectively.
Gelation of globular proteins is induced by the denaturation
unfolding and the subsequent aggregation of the proteins (Doi,
305
1993). Nakamura, Utsumi, and Mori, (1984) proposed the gelation
mechanism of glycinin (11S) protein based on the sucrose density
gradient centrifugation of heated glycinin and TEM as follows: (1)
Within a short time of heating (ca. 15 s) short strands consisting of
about six glycinin molecules, each of which are unfolded but still
keeping a globular shape, are formed (strand I). (2) Strand I asso-
ciates with itself to form a longer straight strands (strand II). (3)
Strand II associates with itself and/or with strand I to form both
branched and unbranched strands (strand III). The gel network
could then form from these strand III units (Fig. 7). Renements of
this mechanism were proposed later, but the essential feature have
not been refuted.
Although interest originally arose in the context of disease
states, amyloid bril formation has attracted much attention in the
food, pharmaceutical and related area because of its potential usage
as a texture modier (van der Linden, 2012; Pearce, Mackintosh &
Gerrard 2007). For example, (Veerman, Sagis, & van der Linden,
2003) reported that, apart from, for example, actin, brinogen,
and tubulin, also several globular food proteins are able to assemble
into long brils, which is generic for proteins. They found the
protein gels prepared from an extremely low concentration 0.07%
using amyloid brils of b-lactoglobulin.
Tang and Wang (2010) investigated bril formation of 7S, 11S
and their mixture using thioavin T and Congo Red spectroscopic
techniques, far UVeCD and AFM (Fig. 8), and concluded that the
formed aggregates of various soy globulins satisfy all three criteria
for the amyloid bril (Nilsson, 2004) and, thus, can be considered to
be a kind of amyloid-like bril.
The gelation process of biopolymers has been studied exten-
sively by small deformation oscillatory measurements (Nishinari,
1997, 2000). A time evolution of storage and loss shear moduli, G0
and G00 , for soymilk in the presence of a coagulant (glucono-d-
lactone, GDL hereafter) has been reported (Fig. 9) (Nishinari et al.
1991). This experiment was proposed as a simulation of tofu-curd
making process at 60  C. In usual tofu production soymilk is
heated at higher temperatures, and the gelation rate is much faster
than reported in the rst paper in 1991, and in such a case it was
difcult to determine the gelation time accurately after mixing GDL
with soymilk even if it was done quickly.
Nagano, Mori, and Nishinari (1994a) monitored the storage
modulus G0 during the gelation process of 15% (w/v) 7 S globulin
solution at pH 7.6, heated at 80  C for 30 min, and then cooled to
20  C at 2  C/min, and then heated to 80  C at the same rate (Fig. 10).
The storage modulus G0 increased at 80  C for 30 min, which in-
dicates that hydrophobic interaction is important for gel formation.
Subsequent increase in G0 on cooling from 80  C to 20  C and the
decrease in G0 on heating from 20  C to 80  C indicates the contri-
bution of hydrogen bonding to the network formation (Fig. 10).
Fig. 7. Formation of soluble aggregates during the gelation of glycinin (Nakamura
et al., 1984).
306 K. Nishinari et al. / Food Hydrocolloids 39 (2014) 301e318
Fig. 8. AFM images of the formed amyloid-like brils of soy b-conglycinin (A), glycinin
(B), and their 1:1 mixture (C). The brils were obtained by heating at 80  C for a period
of 12 h. The height proles of the brils located at red lines (within the gures) are
presented below the AFM images. (Tang & Wang, 2010). (For interpretation of the
Fig. 9. Storage and loss shear moduli of soymilk in the presence of GDL at 60  C. The
gelation time to is set as the origin of the coordinate for the simplicity. (Nishinari et al.
1991).
Nagano, Akasaka, and Nishinari (1994) showed that the plateau
region in mechanical spectra for 15 wt% glycinin began to appear
only above the denaturation temperature about 80  C (Fig. 11). A
similar phenomenon was also observed for b-conglycinin; G0 began
to show the plateau only above 65  C. Nagano et al. (1994) obtained
difference spectra of FTIR absorption for glycinin and b-conglycinin
by subtracting the spectrum of sol from that of gels, which were
formed by heating at 85 and 75  C, respectively. The absorption at
1618 cml was assigned to the formation of b-sheet structures. The
absorbance decrease in the amide I0 or l region was correlated well
with the degree of denaturation. The bands at both 1680 and
1618 cml are characteristic of the b-sheet structure.
As shown in Fig. 12, the percentage absorbance change at
1618 cml as a function of heating temperature for glycinin and b-
conglycinin began to increase above their denaturation tempera-
ture, 65 and 80  C, respectively, and was found to be consistent
with the change in mechanical spectra induced by heating.
Fig. 13 shows the good correlation between storage modulus G0
at 0.87 Hz and the absorbance change at 1618 cm1, indicating that
the gel structure is strengthened through the formation of b-sheet.
On the contrary, Wang and Damodaran (1991) and Mills, Huang,
Noel, Gunning, and Morris (2001) reported that b-sheet decreased
with increasing temperature based on the CD measurement.
Whether the b-sheet decreased or increased with increasing tem-
perature might depend on the concentration used for FTIR and CD.
The protein concentration was 10% in the former, while 0.015% in
the latter. Nagano, Akasaka, and Nishinari (1995) reported CD using
0.02% b-conglycinin solution, and found that b-sheet decreased
with increasing temperature. They interpreted that the presence of
many intermediate states between native and denatured states
should be taken into account in the analysis, which was in accor-
dance with (Iwabuchi, Watanabe, & Yamauchi, 1991) who sug-
gested that b-conglycinin molecules could assume many different
denatured conformations depending on temperature above the
denaturation temperature.
Diffusive wave spectroscopy (DWS) was recently introduced to
detect the structural change of the aggregates in turbid systems
(Alexander & Dalgleish, 2004). Although the photon transport
mean free path and the particle size observed by this method is not
still established, Nik, Alexander, Poysa, Woodrow, & Corredig (2011)
and Ringgenberg, Alexander, & Corredig (2013) found, on lowering
pH, the steep increase in the photon transport mean free path and
the particle size earlier than the rise in storage modulus by small
references to colour in this gure legend, the reader is referred to the web version of
this article.)
 K. Nishinari et al. / Food Hydrocolloids 39 (2014) 301e318
Fig. 10. Gelation process of 15% (w/v) 7 S globulin solution at pH 7.6. The solution was
heated at 80  C for 30 min, cooled to 20  C at 2  C/min, and then heated to 80  C at
2  C/min. (Nagano et al., 1994a).
deformation rheology in their study on soymilk (Fig. 14), and sug-
gested that DWS is more sensitive to changes at micromolecular
level than rheology, as small oscillatory measurements still apply
some strain to the samples. It is difcult to know whether the
structural formation is affected by the strain even though it is very
small. DWS may be useful to understand the initial stage of the
gelation, but does not seem so effective to analyse the later stage
and the resulted gel, and its further development is expected.
3.2.1. Effect of pH on gelation
It has been reported that the ionized carboxyls have an ab-
sorption band around 1570 cm1, whereas protonated carboxyls are
characterized by a band near 1710 cm1. Nagano, Mori, and
Nishinari (1994b) found that the different spectra at acidic pHs
with the spectrum at pH 7.0 indicated that the ionized carboxyls
had an absorption band at 1562 cm1 while protonated carboxyls
were characterized by a band at 1718 cm1 respectively. The degree
of protonation of the carboxyl groups of b-conglycinin was shown
to increase with decreasing pH below pH 6.0.
Fig. 11. Frequency dependence curves of G 0 at 20  C for 15wt% glycinin in 35 mM
phosphate buffer, pH 7.6, which was heated at various temperatures for 30min. (B)78,
(:)80, (,)83, (C)85,and (6)90  C. (Nagano, Akasaka, et al., 1994).
307
Fig. 12. Temperature dependence of absorbance change for glycinin and b-conglycinin
in FTIR at 1618 cml (Nagano, Akasaka, et al., 1994).
FTIR spectra for b-conglycinin gels showed that a band at
1620 cm1 (associated with exposed b-strands) developed with
decreasing pH suggested that b-conglycinin undergoes denatur-
ation with increasing protonation of its carboxyl groups, which
leads to an increase in the amount of exposed b-strands. The
exposed b-strands then intermolecularly bond to form gel net-
works. This process is promoted with decreasing pH, and as a result,
rigid gels form at acidic pH values.
Nagano et al. (1994b) showed that denaturation peak temper-
ature of b-conglycinin detected in heating DSC shifted to lower
temperatures with decreasing pH. Since the denaturation is the
prerequisite of gel formation for globular proteins, it is consistent
with the experimental nding that the storage modulus increased
with decreasing pH (Fig. 15). A similar experimental nding that the
storage modulus increased with decreasing pH was shown for
Fig. 13. Relationship between the storage modulus G0 at 0.87 Hz and the absorbance
change at 1618 cml (Nagano, Akasaka, et al., 1994).
308 K. Nishinari et al. / Food Hydrocolloids 39 (2014) 301e318

Fig. 15. The gelation process of 15% (w/w)b -conglycinin dispersions in the presence of
2.5 % NaCl at various pHs at 80  C. 6 :pH 3.3; , :pH 3.5; - :pH 3.8; : :pH 4.3; B:pH
5.5; C :pH 7.3 (Nagano et al., 1994b).

with increasing ionic strength by DSC measurements. Similar ef-

fects of salt on the gelation can be found for 7w/w% b-lactoglobulin,
an important milk whey protein, in solutions of sodium chloride
and calcium chloride (Foegeding, Kuhn, & Hardin, 1992). The in-
crease of the storage modulus on heating at 80  C was found to
delay with increasing ionic strength. It was also shown that the
storage modulus of b-lactoglobulin continued to increase after the
temperature was lowered from 80  C to 25  C. Nagano et al. (1994a)
also showed that the gelation of b-conglycinin was retarded with
increasing added guanidine hydrochloride (GuHCl) and that no
gelation occurred in the presence of 4 M GuHCl.

3.2.3. Effect of protein concentration on the elastic modulus of soy

Fig. 14. Gelation behaviour of soymilk containing 0.8% (open symbols) or 1.6% (lled
symbols) GDL. A) Storage modulus (G0 ) obtained from rheology data; B) apparent
radius of particles and C) turbidity (1/l*) parameter obtained from DWS measurements
of 4% protein soymilk (circles) and 7% protein soymilk (triangles). Representative runs
of four replicate experiments. (Ringgenberg et al., 2013).
gels
It has been known empirically that the elastic modulus of gels E
is proportional approximately to the square of the concentration C
of gelling polymers. It has more generally been expressed by a
percolation model: E k [(C e C0)/C0)] n, where C0 is the minimum
polymer concentration required for gel formation (Nishinari, 2000).
The exponent n close to two for various polymers is reported, but
the reported value of n for soybean protein is much larger than 2.

12 wt% SPI in the presence of 0.2 M NaCl (Renkema, Gruppen, & van
Vliet, 2002). They also found that denaturation peak temperature of
b-conglycinin in SPI detected in heating DSC shifted to lower
temperatures with decreasing pH.

3.2.2. Effects of salts on gelation

The same rheological measurement showed that sodium chlo-
ride retards the gelation process of b-conglycinin. Nagano et al.
(1994a) found that the gelation time became longer and the gela-
tion rate became slower for b-conglycinin with increasing NaCl
concentration by measuring the complex shear moduli at 80  C
(Fig. 16). This nding is consistent with the stabilization of protein
molecules by sodium chloride to heat denaturation. Wongprecha,
Takaya, Kawase, Nagano, and Nishinari (2000) showed that the
DSC endothermic peak accompanying denaturation of soy glycinin
shifted to higher temperatures with increasing concentration of Fig. 16. Gelation process of 15% (w/v) 7s globulin solution in the presence of NaCl at
added NaCl. Lakemond et al. (2003) also found that the denatur- various concentrations at pH 7.6. The solution was heated at 80  C for 30 min and
ation temperature of soy glycinin shifted to higher temperature cooled to 20  C at 2  C/min (Nagano et al., 1994a).
 K. Nishinari et al. / Food Hydrocolloids 39 (2014) 301e318
The exponent n 1.7 was obtained for 11S soy globulin in the
presence of GDL from the paper by Kohyama and Nishinari (1992),
and this is intermediate between the values 1.56 (pH 3.8,
NaCl 200 mM) and 2.04 (pH 7.6, NaCl 200 mM) reported by
van der Linden and Sagis (2001) who recalculated the exponent
using values reported by Renkema & van Vliet (2004). Kohyamas
data of the saturated storage moduli were measured around
pH 4. The exponent 1.7 is close to the value for ovalbumin (1.56e
1.80) and casein (1.93e2.0) tabulated by van der Linden et al. These
authors stated that using a percolation model rather than using a
fractal model where C0 is assumed to be zero is more reasonable to
establish a generic description.
In the gel formation of globular proteins, some portion which is
not completely denatured may be incorporated in gel network, and
in such a situation the effective concentration contributing to the
elasticity may be lower than the total protein concentration. This
should be taken into account to discuss the concentration depen-
dence of elastic modulus of globular protein gels.
3.2.4. Transparency of soy gels
Most vegetable protein gels are turbid. Some trials to make a
transparent gel from plant source to replace gelatin have been
done. Utsumi, Nakamura, and Mori, (1982) found that the turbidity
of the gels decreased with increasing protein concentration. In fact,
at 20% protein concentration, the absorbance at 600 nm of a gel
formed by heating for 20 min was less than 0.01 and could be
judged macroscopically to be transparent. Thus, they concluded
that the higher the protein concentration, the more transparent the
gel. Hatta, Kitabatake, and Doi (1986) reported that by heating
ovalbumin at various pH and ionic strengths, a transparent solu-
tion, a transparent gel, a turbid gel, a turbid suspension of oval-
bumin could be obtained (Fig.17).
Bacon, Noel, and Lambert (1990) have found that the formation
of clear gels was favoured under conditions of high net charge
when the electrostatic repulsive forces between protein molecules
Fig. 17. Hardness and turbidity of a heat-induced gel from ovalbumin at various pHs.
The 5% (w/v) ovalbumin solution with 20mM NaCl was heated at 80  C for 1 h B-B,
hardness; C-C, turbidity (absorbance at 600 nm); X - X, electric conductivity (Hatta
et al., 1986).
309
can be maintained, thereby minimizing the formation of large ag-
gregates. These conditions are typically realized at low salt con-
centration and at pH values away from the isoelectric point of the
protein. They examined the turbidity of gels of vegetable proteins
and found that the transparency of gels of 10% French bean protein
was almost close to that of a 4% gelatin gel and the melting point
was much higher (90  C) than that of a 4% gelatin gel (32  C). This
10% French bean protein was shown to be very transparent when
the concentration of the added sodium chloride was below 1%.
Their soybean protein gels showed a lower transparency but a
higher melting point (92  C). They suggested possible application of
vegetable protein gels for vegetarian substitutes in products that
traditionally use gelatin as a setting agent. These uses include table
jellies, powdered avoured mixes and stabilizing the juices in pies
and canned meat products.
Guo, Zhang, and Yang (2012) recently proposed a new two-step
method for preparing food globular protein-based hydrogels with
transparent appearance and proper mechanical strength. Dextran
sulfate (DS) was rst introduced to alter the heat-induced aggre-
gation of glycinin which was extracted from soy akes. The elec-
trostatic complexes that consisted of glycinin and DS acted as the
building blocks for the hydrogel. Microbial Transglutaminase
(MTGase) was used to cross-link the electrostatic complexes in the
second step. They examined the inuence of DS/glycinin ratio and
ionic strength on the gelation kinetics and mechanical properties
using small and large deformation rheological analysis, and
observed the structure by SEM. Fig. 18 shows the phase diagram of
MTGase induced glycinin/DS gel as a function of sodium chloride
versus DS concentration.
3.2.5. Effect of coagulants on gels
Most commonly used coagulants to make tofu are glucono-
delta-lactone (GDL) and calcium sulphate (CaSO4). Kohyama, Sano,
and Doi (1995) studied the gelation of SPI using these coagulants,
and found that the gelation by calcium was faster than by GDL
(Fig. 19) and the main molecular forces are hydrophobic interaction.
Fig. 18. Phase diagram of sodium chloride versus DS concentration for MTGase
induced glycinin/DS gel at pH 7.0. The dispersions were heated at 95  C before gelation.
Weak turbid gels, T < 1%; Turbid gels T < 1%; Translucent gels, 1% < T < 10%;
Transparent gels, T > 10%. (Guo, Zhang, et al., 2012).
310 K. Nishinari et al. / Food Hydrocolloids 39 (2014) 301e318
Fig. 19. Typical gelation curves for 5% SPI at 70  C. (Kohyama et al., 1995).
The value of the saturated value of the storage modulus for 5% SPI
increased with CaSO4 concentration up to 35 mM and then
decreased while the rate constant k of the gelation continued to
increase at higher concentration than 35 mM (Fig. 20). This accel-
erating gelation action of the coagulant calcium sulphate (CaSO4)
was similar as observed with GDL (Kohyama & Nishimari, 1992;
Kohyama & Nishinari, 1993; Yoshida, Kohyama, & Nishinari, 1992).
Higher concentrations of CaSO4 than 35 mM led to larger aggregates
and syneresis of the gels; therefore, the apparent value of the
saturated storage modulus decreased. They proposed a gelation
mechanism of SPI in the presence of these coagulants; after the
exposure of hydrophobic groups by heat denaturation soluble ag-
gregates are formed, and then the protons from GDL or calcium ions
from CaSO4 shields the electrostatic repulsion of negative charges
on the surface of soluble aggregates and leads to the network for-
mation. Essentially both play a similar role in spite of kinetic dif-
ference, and it is consistent with the microscopic observation which
shows a similar microstructure (Deman, Deman, & Gupta, 1986). Liu
and Kuo (2011) recently found similar results for GDL and CaSO4
although they found that the network of SPI gels prepared with GDL
was more compact and smoother than those with CaSO4. They
attributed the difference to the rapid dissociation of CaSO4 mole-
cules at low temperature and a shorter gelation time.
Fig. 20. Dependence of G0sat and rate constant of gelation k of 5% SPI on CaSO4 con-
centration at 70  C. (B) G0sat ; (,) k. (Kohyama et al., 1995).
Nagano examined the difference of fracture behavior and
structure of tofus prepared by MgCl2 and GDL (Nagano, 2009). He
found that networks of tofu using soy Fukuyutaka prepared by GDL
is much denser than that prepared by MgCl2 as shown in Fig. 21.
Since the gelation rate is much faster when magnesium chloride is
used than when GDL is used, the processing is easier for GDL and
also with less syneresis. Therefore, he could get tofu enough rm
which was self-standing with 7 times water by using magnesium
chloride and 14 times water by using GDL.
3.2.6. Transglutaminase induced gelation of soybean globulins
Although the catalyzing action of transglutaminase(R-glutamyl-
peptide:amine g-glutamyl-transferase) on the acyl transfer reaction
between protein bound glutamyl residues and primary amines has
been known for a long time, it was not much used in food industry
until the Ajinomoto group proposed to use it as a crosslinking agent
for various proteins including soybean globulins (Motoki & Seguro,
1998; Nio, Motoki, & Takinami, 1986). The crosslinking reaction of
TGase for soyglobulins was studied using SDS-PAGE (sodium
dodecyl sulphate-polyacrylamide gel electrophoresis) by many
groups and it is widely accepted that A and B subunits in glycinin as
well as a0 , a and b subunits in b-conglycinin are crosslinked after
30 min of incubation (Tang, Wu, Chen, & Yang, 2006; Yasir, Sutton,
Newberry, Andrews, & Gerrard, 2007).
Nonaka et al. (1996) reported that the incubation of soymilk
with TGase before GDL induced gelation will make tofu more
retort-resistant, i.e., suppression of water release, and hardening of
tofu pieces packed with water in a retort-pouch (Fig. 22). They
found that soy proteins have been polymerized by TGase.
3.2.7. Cold-set gels of soy protein
Dumoulin, Ozawa, and Hayashi (1998) and Molina, Defaye, and
Ledward (2002) reported the high-pressure-induced (>300 MPa)
gel formation of soy protein at high protein concentrations, such as
17% w/w and 20% w/v, respectively. At low protein concentration
the dispersions did not form a self-supporting gel. These gels were
shown to have high water holding capacity (>80%) but their me-
chanical strength was lower than that of heat-induced gels at the
same concentration, indicating that the denaturation induced by
high pressure and heat was different. Hydrogen bonding is known
to be strengthened by pressure, but as for the effect of pressure on
the hydrophobic interaction seems to be still a matter of debate.
Molina et al. (2002) reported that an endothermic peak accompa-
nying the denaturation of b-conglycinin disappeared after treating
at 400 MPa whereas pressure treatment up to 400 MPa or heat
treatment up to 90  C had no effect of denaturation of glycinin, as
shown by their DSC peak. On the contrary, Puppo et al. (2004)
found that b-conglycinin showed a smaller endothermic peak in
heating DSC even after treatment at 600 MPa, whereas is glycinin
completely denatured by the same pressure treatment and showed
no DSC peaks. Whether this discrepancy is caused by the difference
in the protein concentration (Molina et al. used 20% while Puppo
et al. 5%) or by other reasons has not yet been claried.
Maltais, Remondetto, Gonzalez, and Subirade (2005) found that
heat-denatured soy protein dispersions with lower concentration
(6%e9%) than critical concentration for gelation could form a gel by
adding calcium chloride. Above the critical concentration, the heat
denaturation leads to gelation as described above. They used cal-
cium induced cold-set soybean gels as a delivery tool of a heat labile
nutraceutical compound riboavin (Maltais, Remondetto, &
Subirade, 2009). They prepared two different gels of 9w/w SPI by
changing calcium concentration, lamentous (10 mM CaCl2) and
particulate (20 mM CaCl2), and examined the swelling behaviour of
these gels in simulated gastric uid (SGF) and simulated intestinal
uid (SIF). They found that gels swelled in SGF while they rst
 K. Nishinari et al. / Food Hydrocolloids 39 (2014) 301e318
Fig. 21. CLSM photos of tofus prepared by MgCl2 (left) and GDL (right). (Image size:127.3  127.3 mm) (Nagano, 2009).
swelled but later they shrunk and then collapsed in SIF. They
correlated this swelling behaviour with release of riboavin and
concluded that calcium induced cold-set soybean gels could be
used as vehicles for entrapping bioactive molecules to be delivered
and absorbed in the intestines (see Fig. 23).
Speroni and Anon (2013) reported high pressure denatured
dispersions of SPI, a b-conglycinin enriched fraction (7SEF) and a
glycinin enriched fraction (11SEF) with lower concentration also
formed self-standing cold-set gels by subsequent calcium incor-
poration, and suggested the possibility of incorporating heat-labile
compounds or probiotics during the gelation step. 7SEF formed
aggregated gels with low water holding capacity whereas 11SEF did
not form self-standing gels. SPI formed the better gels: ordered and
with high water holding capacity. It is necessary to take into ac-
count the different sensitivity of each protein to high pressure
treatment. Denaturation of each protein by high pressure treat-
ment is different from that by heat treatment; 11S is 100% dena-
tured after a 400 MPa treatment, while 7S conserves about a 30% of
native structure after a 600 MPa treatment. This is in line with
Puppo et al. (2004) but contradicts Molina et al. (2002).
Fig. 22. Retort-induced weight decrease of tofus with different transglutaminase
concentration. The reaction time was 30 min (Nonaka et al., 1996).
311
3.2.8. How to enhance the gelling ?
The combination of rigid chains and exible chains to make a
strong gel has been studied for creating new strong materials
(Gong, 2010). Since it was shown that the brils can be formed from
soyproteins (Akkermans et al., 2007; Tang et al, 2010), the combi-
nation of soy brils with less rigid food polymer chains is expected
to be used to create a wide range of various textures as has been
done already for the bril from b-lactoglobulin and k-carrageenan
(Jones, Adamcik, Handschin, Bolisetty, & Mezzenga, 2010).
Frozen and dried tofu has been used extensively in Japanese
cooking because of the possibility of the longer storage time. After
freeze-drying tofu became sponge-like and the smooth texture is
lost, therefore, the better processing was investigated by many
research groups. Fig. 24 shows the SEM of frozen and dried tofu
frozen at various temperatures. With decreasing freezing temper-
ature, the pore size of the network became smaller as expected
because the temperature was lowered faster when tofu was put at
lower temperature thus passing faster the temperature range of
maximum ice crystal formation.
Nakao, Yamaguchi, and Taguchi (1994) found an optimal for-
mula for preparing a freezable tofu using curdlan 1.2% and waxy
corn starch 3%. This freezable processed tofu was used in frozen
foods such as Mapo-dofu (a Chinese dish of soybean curd with spicy
minced meat), Agedashi dofu (deep-fried tofu in stock) and Miso
soup, and resulted in the smooth and soft texture. They prepared
Fig. 23. Impact of gastrointestinal conditions on riboavin release from lamentous
and particulate gels (Maltais et al., 2009).
312 K. Nishinari et al. / Food Hydrocolloids 39 (2014) 301e318
Fig. 24. Left SEM of freeze-dried tofu: A, 35  C freezing; B, 20  C freezing; C, 5 C freezing, Right SEM of freeze-dried tofu: A, conventional freeze-dried tofu; B, freeze-dried tofu
with curdlan (Nakao et al., 1994).
also freeze-dried tofu by an optimal formula of curdlan 0.5%, waxy
corn starch 1% and a controlled freezing condition (Fig. 24 right).
Jambrak, Lelas, Mason, Kresic, and Badanjak (2009) reported
that the solubility of soy proteins is increased after ultrasound
treatment, which was attributed to the unfolding and breaking of
peptide bonds by hydrolysis. Hu, Fan, et al. (2013) reported that
high intensity ultrasonic pre-treatment (HUS) of SPI improved the
water holding capacity and gel strength of GDL-induced-SPI-gels
(GISG). They showed that HUS pre-treatments reduced particle
size, increased surface hydrophobicity of SPI and formed soluble
aggregates, leading to denser and more uniform GISG, and thus the
potentiality in food industry.
Some research groups reported that acetic acid bacteria
fermentation reduce the beany avour which limits the use, and
tried to use soymilk for the production of yogurt-like products. Cruz
et al. (2009) compared the yogurt-like product prepared by various
methods; after treating by autoclaving (AC), ultrahigh temperature
(UHT) and ultra high pressure homogenization (UHPH) the soymilk
was fermented by Streptococcus thermophilus and Lactobacillus
delbruekii subsp. Bulgaricus. They measured the rmness and
waterholding capacity, and observed the structure by CLSM, and
found higher rmness for soy-yogurt from UHPH treated soymilk
(300 MPa, 40  C) than that from conventional heat-treated yogurts.
HUT treated yogurt showed a lower water holding capacity than AC
and UHPH treated yogurt, which was found consistent with CSLM
observation that the network of HUT treated yogurt was coarser
than those of AC and UHPH treated yogurt.
SPI is widely used in brine for injected salt soluble meat gel
products such as ham and roast beef to maintain texture and retain
moisture. It is generally known that the gel strength of polymer gels
decreases with decreasing molecular mass. However, it was shown
that enzyme hydrolysed 7S globulin could increase the gel strength
of salt-soluble meat protein gel in comparison with non-hydrolysed
SPI (Tsumura et al. 2005).
Tofu production without producing residues and waste water,
which are also rich in dietary bre, using whole soybean powder
has attracted much attention. CLSM observation of soybean curd
made from powdered whole soybean and normal tofu with
approximately the same protein concentration showed that the
latter tofu has a ner and a more homogeneous network than the
former curd suggesting that ingredients other than soy globulins
dont contribute to the network structure (Yoshimura, Naito,
Nagano, & Nishinari, 2007).
3.3. Emulsication
3.3.1. Relation between hydrophobicity and emulsifying
Any emulsion is in a non-equilibrium state and cannot last so
long, but even within a limited time scale the emulsifying function is
important and useful in the food industry for production and
storage. In practice, stability is a relative term which depends on the
context. For some food emulsions, such as cake batter or cooked
sauces, the required time scale for stability is only a few minutes or
hours. But for other products, such as soft drinks and cream liqueurs,
emulsion stability must be maintained over a period of several
months or years (Dickinson, 2009a). The emulsifying activity was
dened as the maximum oil quantity which can be emulsied by a
xed amount of the protein, and the emulsion stability has been
often dened operationally by the velocity of phase separation into
water and oil during storage of emulsion (Pearce & Kinsella, 1978).
Globular proteins as emulsier are used mainly to make oil in
water (O/W) emulsions. Since the main role of the emulsier in the
emulsion production is to adsorb at the surface of the freshly
formed ne droplets and so prevent them from coalescing with
their neighbours to form larger droplets again, globular proteins
should be denatured at the interface to cover the droplets. Hy-
drophobic amino acids buried in the core of globular protein should
be exposed and adsorb onto the surface of oil droplets, and the
hydrophilic amino acids should be within aqueous phase acting as a
steric barrier against coalescence and occulation.
Emulsiers must have both hydrophobic and hydrophilic groups
to interact with oil and water. Kato and Nakai (1980); Kato, Osako,
Matsudomi, and Kobayashi (1983) found a good correlation of
emulsifying activity index and emulsion stability with surface hy-
drophobicity of proteins using ovalbumin, soy 7S globulin, k-casein,
b-lactoglobulin, and bovine serum albumin (Fig. 25 and Fig. 26).
The surface hydrophobicity of 7S globulin, ovalbumin and k-casein
was found to increase with heat denaturation, whilst that of b-
lactolobulin and bovine serum albumin decreased. Kato et al.
(1983) found the curvilinear correlation between the foaming po-
wer of proteins and surface hydrophobicity during heat denatur-
ation, and found no signicant correlation between the foam
stability and the surface hydrophobicity of proteins.
Because of the low surface hydrophobicity, large molecular size,
and low molecular exibility, glycinin cannot adsorb rapidly to the
airewater interface (Kinsella, 1979; Wagner & Gueguen, 1995;
Wagner & Gueguen, 1999). Liu, Lee, and Damodaran (1999) iso-
lated acidic subunits from 11S, and found that isolated acidic sub-
units adsorbed to the airewater interface faster than 11S.
Rivas and Sherman (1984) found that 7S formed a stronger lm
at interface than 11S irrespective of pH or addition of NaCl, and
interpreted that the 7S globulin molecules had a greater degree of
intra- and inter-molecular cohesion and so they formed more or-
dered lms. Their view was consistent with generally accepted
concept that molecules with more available hydrophobic residues
develop stronger, and more concentrated, gel-like structures at an
interface, since hydrophobic interactions contribute more rigidity
to the lm.
Wagner and Gueguen (1995) showed that the dissociation,
deamidation and reduction of glycinin led to a decrease in
 K. Nishinari et al. / Food Hydrocolloids 39 (2014) 301e318
Fig. 25. Correlation of emulsifying activity index with surface hydrophobicity of proteins (Kato et al., 1983).
molecular size, and increase in surface hydrophobicity and electric
charge thus improving the emulsifying function.
Kimura et al. (2010) suggested that carbohydrate moieties in 7S
globulin plays an important role in increasing the emulsifying
property based on the comparative study of 7S and 11S globulins
extracted from pea, faba bean, cowpea and French bean.
It has been known that in soybean seeds there are oil bodies
consisting of a triacylglycerol core, which is covered by a layer of
phospholipids and a protein oleosin. One very special characteris-
tics of soy proteins is its high oil holding capacity as is seen when
tofu-curd is cooked in hot water. While fat is exuded out when most
animal meat is cooked in water, no oil is exuded out from tofu-curd.
Guo et al. (1999) stated that lipid incorporation took place by the
conjugation of the lipid and protein particles based on the exami-
nation of lipid in three fractions, oating, particulate, and soluble
fractions obtained by centrifugation. The oil bodies are believed to
exist in naturally emulsied state, and many research groups have
already extracted from soybeans in aqueous environment without
using organic solvent such as hexane, and studied the application in
food products in place of emulsied soybean oil, for example, in
dressings, sauces, dips, beverages, and desserts. Additional advan-
tages of using natural soybean oil bodies in foods, rather than
emulsied bulk soybean oil, are that neither emulsiers nor ho-
mogenization procedures are required (Chen & Ono, 2010; Iwanaga
et al., 2007).
Lam and Nickerson (2013) have recently written a review on
protein-stabilised emulsions, and Evans, Ratcliffe, and Williams
Fig. 26. Correlation of emulsifying stability index with surface hydrophobicity of proteins (Kato et al., 1983).
313
(2013) have written a review on emulsion stabilisation by protein-
polysaccharide complexes.
3.3.2. Emulsions stabilized by soy proteins
Keerati-u-rai and Corredig (2009) studied effects of heating on
the SPI stabilized soybean oil-in-water emulsion. The droplet size
distribution was monomodal when the 10% oil emulsion was pre-
pared by unheated 1% SPI (Fig. 27). They examined the effects of
changing the order of the process (heating the solution before
emulsication, or heating the emulsion) on the droplet size dis-
tribution, and found the smaller droplet size for emulsions pre-
pared by heated SPI solutions than emulsions heated after
emulsication, but the amount of protein necessary to stabilize the
emulsion was higher because the adsorbed layers on oil droplet
consisted of aggregated proteins. This hypothesis was proved by
increasing concentration of SPI to 2%; the population of larger
droplet size around 10 micron in the emulsion prepared by 1% SPI
shifted to smaller sizes (Fig. 27).
They found an endothermic peak at around 95  C in addition to
that at 68  C caused by the denaturation of b-conglycinin and at
85  C by glycinin, and suggested the structural change of the pro-
tein upon adsorption on the oil droplet. They found that all the
protein subunits to be present at the interface in an aggregated
form in SDS-PAGE when the solutions were heated before emulsi-
cation. However, they found that b subunit of b-conglycinin and
basic subunit of glycinin disappeared in SDS-PAGE after heating at
95  C, and suggested that heat-induced complexes were formed
314 K. Nishinari et al. / Food Hydrocolloids 39 (2014) 301e318
Fig. 27. Droplet size distribution of 10% oil stabilized by 1% SPI. The emulsions were
prepared with heated solutions (empty symbols) or unheated solutions and then
heated after homogenization (lled symbols). Unheated control (C); 75  C for 15 min
(,, -); 95  C for 15 min (6, :). (Keerati-u-rai & Corredig, 2009).
and remained soluble as found previously by Damodaran and
Kinsella (1982) when heating was applied after emulsication.
Partial protein hydrolysis to enhance their emulsifying proper-
ties which had been used for milk and wheat proteins were applied
for soy proteins. Tsumura et al. (2005) modied the structure of
soyproteins by enzyme degradation, and succeeded in getting
reduced-b-conglycinin hydrolysate and reduced-glycinin hydroly-
sate. While the control unmodied soy protein shows a high EAI at
neutral pH, its EAI is very poor at acidic pH, both reduced-b-con-
glycinin hydrolysate and reduced-glycinin hydrolysate show a high
EAI at acidic pH (Fig. 28) indicating that modication of the protein
structure by enzyme will lead to an improved EAI (Tsumura et al.
2005).
Chen, Chen, Ren, and Zhao (2011) improved emulsifying capa-
bility of SPI using combined extrusion pre-treatment and controlled
enzymatic hydrolysis, and they attributed it to the increased protein
solubility and decreased molecular weight. Although hydrolysed
globulins can migrate faster to the interface which is advantageous
to improve the emulsication, there should be an optimum degree
of hydrolysis; indeed it was shown that highly hydrolysed sunower
protein isolate tended to saturate the continuous phase rather than
adhere to the watereoil interface (Conde & Patino, 2007).
3.3.3. How to enhance the emulsication?
Jambrak et al. (2009) reported that both emulsifying and foam-
ing ability of soy proteins can be increased by ultrasound treatment.
Fig. 28. EAI of the hydrolysates of SPI at various pH values (Tsumura et al., 2005).
Ultrasound treatment with 20 kHz probe increased the solubility of
soy protein concentrates, the specic surface area and EAI.
Chove, Grandison, and Lewis (2007) showed a possibility to
modify the structure by microltration. Fractionation was carried
out on SPI produced by isoelectric precipitation of a crude protein
extract, and retentates and permeates were obtained. Emulsions
stabilised by the retentates exhibited higher emulsion stability in-
dex (ESI) and emulsifying activity index (EAI) than those stabilised
with permeates. They also found that the fractions exhibiting high
functionality in terms of solubility, foaming and emulsifying
properties were also richer in 7S globulin soy protein subunits
based on SDS-PAGE.
Wan, Wang, Wang, Yang, and Yuan (2013) attempted to improve
the stability of soy protein isolate (SPI)-based emulsions by incor-
poration of resveratrol (RES) as a natural antioxidant with stevioside
(STE). STE was introduced to enhance the solubility of RES. The water
solubility of RES was shown to increase with increasing STE con-
centration up to its critical micelle concentration, which was ascribed
to the solubilization of hydrophobic RES in STE self-assembled mi-
celles. Wan et al. (2013) showed that STE micelles competitively
adsorbed at the oilwater interface with SPI, forming a mixed SPI and
STE interfacial layer, thus resulting in a decrease in particle size and
evident enhancement in the physical stability of SPI based emulsions.
Chen, Zhao, Sun, Ren, and Cui (2013) examined the effect of
oxidation on the emulsifying properties of soy protein isolate, and
reported that emulsions stabilized by moderately oxidized SPI had
a smaller droplet size and better thermal stability in comparison
with the control and over-oxidized SPI.
Although soybean oil bodies have great advantages as previ-
ously noted, they are unstable to aggregation because of the rela-
tively weak electrostatic repulsions between them over a wide
range of pH values (3 < pH < 7), and the stabilization of oil body
emulsions was studied by coating with pectin (Iwanaga, Gray,
Decker, Weiss, & McClements, 2008) or carrageenans (Wu et al.,
2011). Iwanaga et al. (2008) showed that citrus pectin-coated oil
bodies had similar or improved stability compared to uncoated oil
bodies. Wu et al. (2011) using k, i, l- carrageenans to see the effect
of electric charge and conformation on the stabilizing effect, and
found that i-carrageenan was most effective and attributed it to the
most densely charged helical structure, and thus most effective at
creating highly charged interfacial membranes.
Matemu, Kayahara, Murasawa, Katayama, & Nakamura (2011)
found that the emulsifying activity (EAI) and emulsion stability
(ES) of 7S and 11S were signicantly improved upon acylation with
saturated fatty acids. Covalent attachment of fatty acids resulted
into 1.4e2.2 and 1.1e1.8-fold increase in the oil binding capacity of
7S and 11S respectively. Acylated 7S showed 3.0e9.4-fold increase
in the water binding capacity, with no change in acylated 11S. The
surface hydrophobicity of 7S was signicantly improved by acyla-
tion; no changes were observed in the acylated 11S.
Since a pioneering study of Kato, Murata, and Kobayashi, (1988)
on Maillard type reaction between ovalbumin-dextran, extensive
works on protein-polysaccharide conjugates for improving emul-
sication, especially at low pH around 4 where the solubility of
soybean protein shows the minimum, have been reported
(Dickinson 2009b; Evans et al. 2013). Achouri et al. (2010) prepared
glycated CaCl2e11S using k-carrageenan, and noticed the
improvement of EAI and foaming properties at a moderate degree
of glycation. A complex of SPI and chitosan (CS) prepared at 121  C
and at low pH showed an improved EAI at pH4 (Yuan et al. 2013). It
was suggested that physicochemical and functional properties of
SPI could be modulated by heated complexing with chitosan, using
appropriate mixing ratio, molecular weights and charge densities
of chitosan. They found that the droplet size became minimum at
the mixing ratio CS/Glycinin 0.1 (Fig. 29).
 K. Nishinari et al. / Food Hydrocolloids 39 (2014) 301e318
Fig. 29. Effect of CS/glycinin mixing ratio on zeta-potential and droplet size (D43) of
glycinin/CS mixed emulsions at pH 4.5. (Yuan et al., 2013).
Soy whey protein isolate (SWPI)-fenugreek gum conjugates were
prepared by Maillard type reactions in a controlled dry state, and it
was shown that the conjugates had better emulsifying properties
even near the isoelectric pH of protein where the proteins are prone
to aggregate, and could destabilize the emulsion. Fenugreek gum
was partially hydrolyzed using 0.05 M HCl at 90  C for 10 min, 30 min
and 50 min. The ability of the conjugates in lowering the particle size
of emulsions was the highest in the conjugate with unhydrolyzed
fenugreek and it became less effective with increasing time of hy-
drolysis as shown in Fig. 30 (Kasran, Cui, & Goff, 2013). Heating so-
lutions of the conjugates prior to emulsication was shown to
improve their emulsication properties.
Soy soluble polysaccharide (SSPS) has been shown to be a good
emulsier in the food industry (Nakamura, Yoshida, Maeda, &
Corredig, 2006), and it was shown recently to be able to prevent
the destabilization of SPI dispersions and SPI-based oil-in-water (O/
W) emulsions under acidic conditions (Tran & Rousseau, 2013).
3.3.4. Emulsion gels
Emulsion gels were reviewed recently by Dickinson (2012).
Since many processed foods can be regarded as protein-based oil-
Fig. 30. Comparison of average droplet size (d43) for emulsions of canola oil (10 vol%
oil, 0.8% emulsier) stabilized by SWPI-fenugreek conjugate as a function of storage
time from 0 to 28 days at pH 4.0 and stored at 25  C. The time upper right represents
the time of hydrolysis of fenugreek (Kasran et al. 2013).
315
in-water emulsions which can be converted into soft-solid-like
materials by common food processing operations such as heating,
acidication, and enzyme action, there have been many studies on
emulsion gels. Matsumura, Kang, Sakamoto, Motoki, and Mori
(1993) studied the effect of oil content and droplet size on rheo-
logical properties of emulsion gels of 11S globulin. They found that
both storage and loss moduli of the gels increased with increasing
oil content up to 15% oil and with decreasing droplet size. Kim,
Renkema, and van Vliet (2001) conrmed similar results
including large deformation behaviour; both breaking stress and
strain increased with increasing oil content (<volume fraction 0.3)
and with decreasing droplet size.
Tang and Liu (2013) reported the preparation and character-
ization of cold, gel-like soy protein emulsions consisting of un-
treated and preheated SPI at a protein concentration of 6% (w/v),
and various oil volume fractions (0.2e0.6) and NaCl concentrations
(0e500 mM) by microuidization. Both untreated and preheated
SPI emulsions were found to behave gel-like rheologically, but the
latter ones were found more gel-like at a comparable oil volume
fractions. Since both steady shear viscosity and storage modulus of
the emulsion increased with increasing oil volume fraction (Fig. 31),
they attributed it to enhanced inter-droplet interactions, and they
conrmed this bridging occulation of oil droplets by CLSM
observation.
They dont call this state a gel because the chain-like connected
oil droplets may be essentially a liquid just like a previously so-
called weak gel which is essentially a liquid, and should not be
called a gel. There must be a critical oil content above which the
emulsion gel should not be called a gel but should be called gel-like
emulsion. Gels will be broken when subjected to a large deforma-
tion while the gel-like emulsion may not be broken but ow
(Djabourov et al. 2013).
Tang, Luo, Liu, and Chen (2013) recently studied
transglutaminase-set soy globulin-stabilized emulsion gels, and
found that the increase in the glycinin/b-conglycinin ratio pro-
gressively increased the gel stiffness, but signicantly decreased the
water holding capacity. Their CLSM showed that increasing glycinin
content led to the formation of emulsion gel network with a more in
homogenous and porous microstructure. They suggested this gel-
like emulsion may be useful as carriers for heat-labile ingredients
with health effects.
Fig. 31. Frequency dependence of storage modulus of SPI-stabilized emulsions at
various oil volume fractions from 0.2 to 0.6. The protein concentration in aqueous
phase was 6.0% (w/v), and the NaCl concentration 300 mM. The heat pretreatment of
SPI was carried out at 95  C for 15 min, prior to the emulsication (Tang & Liu, 2013).
316 K. Nishinari et al. / Food Hydrocolloids 39 (2014) 301e318
4. Concluding remarks
In spite of enormous efforts, the perfect crystal of the main
storage proteins glycinin and b-conglycinin has not yet been ob-
tained, which makes difcult to get a clear understanding of their
behaviour. It is urgently required to establish this, to exploit further
utilisation of soybean proteins. It is also necessary to elucidate the
interaction among main proteins, glycinin, b-conglycinin, and whey
proteins as well as lipids to develop further application of soy beans
because it is not economical to use each extracted/separated pro-
tein fractions in food industry. It will be useful to compare the
functional properties, gelling and emulsication with globular
proteins of the other origins to get more insight. It is also required to
get systematic understanding about the mechanism of the inter-
action between soybean protein and polysaccharides to develop
their further utilisation in food and pharmaceutical area.
Acknowledgements
Authors wish to thank Prof. T. Ono, Prof. T. Nagano, and Dr. K.
Kohyama.
References
Achouri, A., Boye, J. I., Belanger, D., Chiron, T., Yaylayan, V. A., & Yeboah, F. K. (2010).
Functional and molecular properties of calcium precipitated soy glycinin and
the effect of glycation with kappa-carrageenan. Food Research International, 43,
1494e1504.
Adachi, M., Kanamori, J., Masuda, T., Yagasaki, K., Kitamura, K., Mikami, B., et al.
(2003). Crystal structure of soybean 11S globulin:glycinin A3B4 homohexamer.
Proceedings of the National Academy of Sciences of the United States of America,
100, 7395e7400.
Adamcik, J., & Mezzenga, R. (2012). Proteins brils from a polymer physics
perspective. Macromolecules, 45, 1137e1150.
Akkermans, C., Van Der Goot, A. J., Venema, P., Gruppen, H., Vereijken, J. M., Van Der
Linden, E., et al. (2007). Micrometer-sized brillar protein aggregates from soy
glycinin and soy protein isolate. Journal of Agricultural and Food Chemistry, 55,
9877e9882.
Alexander, M., & Dalgleish, D. G. (2004). Application of transmission diffusing wave
spectroscopy to the study of gelation of milk by acidication and rennet. Col-
loids and Surfaces B-Biointerfaces, 38, 83e90.
Bacon, J. R., Noel, T. R., & Lambert, N. (1990). Preparation of transparent protein gels
from plant seeds. Food Hydrocolloids, 3, 491e495.
Badley, R. A., Atkinson, D., Hauser, H., Oldani, D., Green, J. P., & Stubb, J. M. (1975).
The structure, physical and chemical properties of the soy bean protein glycinin.
Biochimica et Biophysica Acta, 412, 214e228.
Bikbov, T. M., Grinberg, V. Y., Schmandke, H., Chaika, T. S., Vaintraub, I. A., &
Tolstoguzov, V. B. (1981). A study on gelation of soybean globulin solutions.
Colloid and Polymer Science, 259, 536e547.
Chen, L., Chen, J. S., Ren, J. Y., & Zhao, M. M. (2011). Modications of soy protein
isolates using combined extrusion pre-treatment and controlled enzymatic
hydrolysis for improved emulsifying properties. Food Hydrocolloids, 25, 887e
897.
Chen, Y. M., & Ono, T. (2010). The mechanisms for yuba formation and its stable
lipid. Journal of Agricultural and Food Chemistry, 58, 6485e6489.
Chen, Y. M., Yamaguchi, S., & Ono, T. (2009). Mechanism of the chemical compo-
sition changes of yuba prepared by a laboratory processing method. Journal of
Agricultural and Food Chemistry, 57, 3831e3836.
Chen, N. N., Zhao, M. M., Sun, W. Z., Ren, J. Y., & Cui, C. (2013). Effect of oxidation on
the emulsifying properties of soy protein isolate. Food Research International, 52,
26e32.
Chiti, F., & Dobson, C. M. (2009). Amyloid formation by globular proteins under
native conditions. Nature Chemical Biology, 5, 15e22.
Chove, B. E., Grandison, A. S., & Lewis, M. J. (2007). Some functional properties of
fractionated soy protein isolates obtained by microltration. Food Hydrocolloids,
21, 1379e1388.
Clark, A. H. (1998). Gelation of globular proteins. In S. E. Hill, D. A. Ledward, &
J. R. Mitchell (Eds.), Functional properties of food macromolecules (2nd ed.) (pp.
77e142). Aspem Publishers.
Conde, J. M., & Patino, J. M. R. (2007). The effect of enzymatic treatment of a sun-
ower protein isolate on the rate of adsorption at the air-water interface.
Journal of Food Engineering, 78, 1001e1009.
Cruz, N. S., Capellas, M., Jaramillo, D. P., Trujillo, A. J., Guamis, B., & Ferragut, V.
(2009). Soymilk treated by ultra high-pressure homogenization: acid coagula-
tion properties and characteristics of a soy-yogurt product. Food Hydrocolloids,
23, 490e496.
Damodaran, S., & Kinsella, J. E. (1982). Effect of conglycinin on the thermal
aggregation of glycinin. Journal of Agricultural and Food Chemistry, 30, 812e
817.
Deak, N. A., Murphy, P. A., & Johnson, L. A. (2006). Fractionating soybean storage
proteins using Ca2 and NaHSO3. Journal of Food Science, 71, C413eC424.
Deman, J. M., Deman, L., & Gupta, S. (1986). Texture and microstructure of
soybean curd (tofu) as affected by different coagulants. Food Microstructure, 5,
83e89.
Dickinson, E. (2009a). In G. O. Phillips, & P. A. Williams (Eds.), Handbook of hydro-
colloids (pp. 23e49). Woodhead Publ.
Dickinson, E. (2009b). Hydrocolloids as emulsiers and emulsion stabilizers. Food
Hydrocolloids, 23, 1473e1482.
Dickinson, E. (2012). Emulsion gels: the structuring of soft solids with protein-
stabilized oil droplets. Food Hydrocolloids, 28, 224e241.
Djabourov, M., Nishinari, K., & Ross-Murphy, S. B. (2013). Physical gels from biological
and synthetic polymers. Cambridge: Cambridge University Press.
Dobson, C. M. (2003). Protein folding and misfolding. Nature, 426, 884e890.
Doi, E. (1993). Gels and gelling of globular proteins. Trends in Food Science &
Technology, 4, 1e5.
Dumoulin, M., Ozawa, S., & Hayashi, R. (1998). Textural properties of pressure-
induced gels of food proteins obtained under different temperatures
including subzero. Journal of Food Science, 63, 92e95.
Evans, M., Ratcliffe, I., & Williams, P. A. (2013). Emulsion stabilisation using
polysaccharide-protein complexes. Current Opinion in Colloid & Interface Science,
18, 272e282.
Foegeding, E. A., Kuhn, P. R., & Hardin, C. C. (1992). Specic divalent cation-induced
changes during gelation of beta-lactoglobulin. Journal of Agricultural and Food
Chemistry, 40, 2092e2097.
Gong, J. P. (2010). Why are double network hydrogels so tough? Soft Matter, 6,
2583e2590.
Guo, S. T., Ono, T., & Mikami, M. (1999). Incorporation of soy milk lipid into protein
coagulum by addition of calcium chloride. Journal of Agricultural and Food
Chemistry, 47, 901e905.
Guo, J., Yang, X. Q., He, X. T., Wu, N. N., Wang, J. M., Gu, W., et al. (2012). Limited
aggregation behavior of beta-conglycinin and its terminating effect on glycinin
aggregation during heating at pH 7.0. Journal of Agricultural and Food Chemistry,
60, 3782e3791.
Guo, J., Zhang, Y., & Yang, X. Q. (2012). A novel enzyme cross-linked gelation method
for preparing food globular protein-based transparent hydrogel. Food Hydro-
colloids, 26, 277e285.
Hashizume, K., & Watanabe, T. (1979). Heat denaturation of soybean proteins 2.
Inuence of heating temperature on conformational.changes of soybean pro-
teins. Agricultural and Biological Chemistry, 43, 683e690.
Hatta, H., Kitabatake, N., & Doi, E. (1986). Turbidity and hardness of a heat-induced
gel of hen egg ovalbumin. Agricultural and Biological Chemistry, 50, 2083e2089.
Hu, H., Fan, X., Zhou, Z., Xu, X. Y., Fan, G., Wang, L. F., et al. (2013). Acid-induced
gelation behavior of soybean protein isolate with high intensity ultrasonic pre-
treatments. Ultrasonics Sonochemistry, 20, 187e195.
Hu, S. D., Liu, H., Qiao, S. Y., He, P. L., Ma, X., & Lu, W. Q. (2013). Development of
immunoafnity chromatographic method for isolating glycinin (11S) from
soybean proteins. Journal of Agricultural and Food Chemistry, 61, 4406e4410.
Iwabuchi, S., Watanabe, H., & Yamauchi, F. (1991). Thermal denaturation of beta-
conglycinin. Kinetic resolution of reaction mechanism. Journal of Agricultural
and Food Chemistry, 39, 27e33.
Iwabuchi, S., & Yamauchi, F. (1987). Electrophoretic analysis of whey proteins
present in soybean globulin fractions. Journal of Agricultural and Food Chemistry,
35, 205e209.
Iwanaga, D., Gray, D., Decker, E. A., Weiss, J., & McClements, D. J. (2008). Stabilization
of soybean oil bodies using protective pectin coatings formed by electrostatic
deposition. Journal of Agricultural and Food Chemistry, 56, 2240e2245.
Iwanaga, D., Gray, D. A., Fisk, I. D., Decker, E. A., Weiss, J., & McClements, D. J. (2007).
Extraction and characterization of oil bodies from soy beans: a natural source
of pre-emulsied soybean oil. Journal of Agricultural and Food Chemistry, 55,
8711e8716.
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,
386e393.
Jones, O. G., Adamcik, J., Handschin, S., Bolisetty, S., & Mezzenga, R. (2010). Fibril-
lation of beta-lactoglobulin at low pH in the presence of a complexing anionic
polysaccharide. Langmuir, 26, 17449e17458.
Jung, S., Murphy, P. A., & Johnson, L. A. (2005). Physicochemical and functional
properties of soy protein substrates modied by low levels of protease hydro-
lysis. Journal of Food Science, 70, C180eC187.
Kasran, M., Cui, S. W., & Goff, H. D. (2013). Emulsifying properties of soy whey
protein isolate-fenugreek gum conjugates in oil-in-water emulsion model
system. Food Hydrocolloids, 30, 691e697.
Kato, A., Murata, K., & Kobayashi, K. (1988). Preparation and characterization of
ovalbumin-dextran conjugate having excellent emulsifying properties. Journal
of Agricultural and Food Chemistry, 36, 421e425.
Kato, A., & Nakai, S. (1980). Hydrophobicity determined by a uorescence probe
method and its correlation with surface-properties of proteins. Biochimica et
Biophysica Acta, 624, 13e20.
Kato, A., Osako, Y., Matsudomi, N., & Kobayashi, K. (1983). Changes in the emulsi-
fying and foaming properties of proteins during heat denaturation. Agricultural
and Biological Chemistry, 47, 33e37.
 K. Nishinari et al. / Food Hydrocolloids 39 (2014) 301e318
Keerati-u-rai, M., & Corredig, M. (2009). Heat-induced changes in oil-in-water
emulsions stabilized with soy protein isolate. Food Hydrocolloids, 23, 2141e
2148.
Kim, K. H., Renkema, J. M. S., & van Vliet, T. (2001). Rheological properties of soy-
bean protein isolate gels containing emulsion droplets. Food Hydrocolloids, 15,
295e302.
Kimura, A., Tandang-Silvas, M. R. G., Fukuda, T., Cabanos, C., Takegawa, Y.,
Amano, M., et al. (2010). Carbohydrate moieties contribute signicantly to the
physicochemical properties of French bean 7S globulin Phaseolin. Journal of
Agricultural and Food Chemistry, 58, 2923e2930.
Kinsella, J. E. (1979). Functional properties of soy proteins. Journal of the American
Oil Chemists Society, 56, 242e258.
Kitamura, K. (Ed.). (2010). All about soybeans. Tokyo: Science Forum.
Kohyama, K., & Nishinari, K. (1992). The effect of glucono-delta-lactone on the
gelation time of soybean 11S-protein e concentration dependence. Food Hy-
drocolloids, 6, 263e274.
Kohyama, K., & Nishinari, K. (1993). Rheological studies on the gelation process of
soybean 7 S and 11 S proteins in the presence of glucono-delta-lactone. Journal
of Agricultural and Food Chemistry, 41, 8e14.
Kohyama, K., Sano, Y., & Doi, E. (1995). Rheological characteristics and gelation
mechanism of tofu (soybean curd). Journal of Agricultural and Food Chemistry,
43, 1808e1812.
Koshiyama, I., Kikuchi, M., & Fukushima, D. (1981). 2S globulins of soybean seeds. 2.
Physicochemical and biological properties of protease inhibitors in 2S globulins.
Journal of Agricultural and Food Chemistry, 29, 340e343.
Kuipers, B. J. H., van Koningsveld, G. A., Alting, A. C., Driehuis, F., Voragen, A. G. J., &
Gruppen, H. (2006). Opposite contributions of glycinin- and beta-conglycinin-
derived peptides to the aggregation behavior of soy protein isolate hydroly-
sates. Food Biophysics, 1, 178e188.
Lakemond, C. M. M., de Jongh, H. H. J., Paques, M., van Vliet, T., Gruppen, H., &
Voragen, A. G. J. (2003). Gelation of soy glycinin; inuence of pH and ionic
strength on network structure in relation to protein conformation. Food Hy-
drocolloids, 17, 365e377.
Lam, R. S. H., & Nickerson, M. T. (2013). Food proteins: a review on their emulsifying
properties using a structure-function approach. Food Chemistry, 141, 975e984.
van der Linden, E. (2012). From peptides and proteins to micro-structure
mechanics and rheological properties of bril systems. Food Hydrocolloids, 26,
421e426.
van der Linden, E., & Sagis, L. M. C. (2001). Isotropic force percolation in protein gels.
Langmuir, 17, 5821e5824.
Liu, H. H., & Kuo, M. I. (2011). Effect of microwave heating on the viscoelastic
property and microstructure of soy protein isolate gel. Journal of Texture Studies,
42, 1e9.
Liu, M. X., Lee, D. S., & Damodaran, S. (1999). Emulsifying properties of acidic
subunits of soy 11S globulin. Journal of Agricultural and Food Chemistry, 47,
4970e4975.
Maltais, A., Remondetto, G. E., Gonzalez, R., & Subirade, M. (2005). Formation of soy
protein isolate cold-set gels: protein and salt effects. Journal of Food Science, 70,
C67eC73.
Maltais, A., Remondetto, G. E., & Subirade, M. (2009). Soy protein cold-set hydrogels
as controlled delivery devices for nutraceutical compounds. Food Hydrocolloids,
23, 1647e1653.
Maruyama, N., Adachi, M., Takahashi, K., Yagasaki, K., Kohno, M., Takenaka, Y., et al.
(2001). Crystal structures of recombinant and native soybean beta-conglycinin
beta homotrimers. European Journal of Biochemistry, 268, 3595e3604.
Maruyama, N., Salleh, M. R. M., Takahashi, K., Yagasaki, K., Goto, H., Hontani, N., et al.
(2002). The effect of the N-linked glycans on structural features and physico-
chemical functions of soybean beta-conglycinin homotrimers. Journal of the
American Oil Chemists Society, 79, 139e144.
Matemu, A. O., Kayahara, H., Murasawa, H., Katayama, S., & Nakamura, S. (2011).
Improved emulsifying properties of soy proteins by acylation with saturated
fatty acids. Food Chemistry, 124, 596e602.
Matsumura, Y., Kang, I. J., Sakamoto, H., Motoki, M., & Mori, T. (1993). Filler effects of
oil droplets on the viscoelastic properties of emulsion gels. Food Hydrocolloids,
7, 227e240.
Mills, E. N. C., Huang, L., Noel, T. R., Gunning, A. P., & Morris, V. J. (2001). Formation
of thermally induced aggregates of the soya globulin beta-conglycinin. Bio-
chimica Et Biophysica Acta-Protein Structure and Molecular Enzymology, 1547,
339e350.
Molina, E., Defaye, A. B., & Ledward, D. A. (2002). Soy protein pressure-induced gels.
Food Hydrocolloids, 16, 625e632.
Motoki, M., & Seguro, K. (1998). Transglutaminase and its use for food processing.
Trends in Food Science & Technology, 9, 204e210.
Nagano, T. (2009). Vegetable protein. Sen-I Gakkaishi, 65, P462eP465.
Nagano, T., Akasaka, T., & Nishinari, K. (1994). Dynamic viscoelastic properties of
glycinin and beta-conglycinin gels from soybeans. Biopolymers, 34, 1303e1309.
Nagano, T., Akasaka, T., & Nishinari, K. (1995). Study on the heat-induced confor-
mational changes of -conglycinin by FTIR and CD analysis. Food Hydrocolloids, 9,
83e89.
Nagano, T., Mori, H., & Nishinari, K. (1994a). Effect of heating and cooling on the
gelation kinetics of 7S globulin from soybeans. Journal of Agricultural and Food
Chemistry, 42, 1415e1419.
Nagano, T., Mori, H., & Nishinari, K. (1994b). Rheological properties and
conformational states of beta-conglycinin gels at acidic pH. Biopolymers, 34,
293e298.
317
Nakamura, T., Utsumi, S., & Mori, T. (1984). Network structure formation in
thermally-induced gelation of glycinin. Journal of Agricultural and Food Chem-
istry, 32, 349e352.
Nakamura, A., Yoshida, R., Maeda, H., & Corredig, M. (2006). Soy soluble poly-
saccharide stabilization at oil-water interfaces. Food Hydrocolloids, 20, 277e283.
Nakao, Y., Yamaguchi, T., & Taguchi, T. (1994). Preparations of freezable processed
tofu and freeze-dried tofu by using curdlan. Nippon Shokuhin Kogyo Gakkaishi,
41, 141e147.
Nik, A. M., Alexander, M., Poysa, V., Woodrow, L., & Corredig, M. (2011). Effect of soy
protein subunit composition on the rheological properties of soymilk during
acidication. Food Biophysics, 6, 26e36.
Nilsson, M. R. (2004). Techniques to study amyloid bril formation in vitro. Methods,
34, 151e160.
Nio, N., Motoki, M., & Takinami, K. (1986). Gelation mechanism of protein solution
by transglutaminase. Agricultural and Biological Chemistry, 50, 851e855.
Nishinari, K. (1988). Food hydrocolloids in Japan. In G. O. Phillips, D. Wedlock, &
P. A. Williams (Eds.), Gums and stabilisers for the food industry, 4; (pp. 373e390).
Oxford: IRL Press.
Nishinari, K. (1997). Rheological and DSC study of sol-gel transition in aqueous
dispersions of industrially important polymers and colloids. Colloid and Polymer
Science, 275, 1093e1107.
Nishinari, K. (2000). Rheology of physical gels and gelling processes. Reports on
Progress in Polymer Physics in Japan, 43, 163e192.
Nishinari, K., Kohyama, K., Zhang, Y., Kitamura, K., Sugimoto, T., Saio, K., et al. (1991).
Rheological study on the effect of the A5 subunit on the gelation characteristics
of soybean proteins. Agricultural and Biological Chemistry, 55, 351e355.
Nonaka, M., Sakamoto, H., Toiguchi, S., Yamagiwa, K., Soeda, T., & Motoki, M. (1996).
Retort-resistant tofu prepared by incubation with microbial transglutaminase.
Food Hydrocolloids, 10, 41e44.
Ono, T., Katho, S., & Mothizuki, K. (1993). Inuences of calcium and pH on
protein solubility in soybean milk. Bioscience Biotechnology and Biochemistry, 57,
24e28.
Pearce, K. N., & Kinsella, J. E. (1978). Emulsifying properties of proteins: evaluation
of a turbidimetric technique. Journal of Agricultural and Food Chemistry, 26,
716e723.
Pearce, F. G., Mackintosh, S. H., & Gerrard, J. A. (2007). Formation of amyloid-like
brils by ovalbumin and related proteins under conditions relevant to food
processing. J Agric. Food Chem., 55, 318e322.
Pearson, A. M. (1983). Soy proteins. In B. J. F. Hudson (Ed.), Developments in food
proteins-2 (pp. 67e108). Essex, England: Applied Science Pub.
Peng, I. C., Quass, D. W., Dayton, W. R., & Allen, C. E. (1984). The physicochemical and
functional properties of soybean 11S globulin e a review. Cereal Chemistry, 61,
480e490.
Puppo, C., Chapleau, N., Speroni, F., de Lamballerie-Anton, M., Michel, F., Anon, C.,
et al. (2004). Physicochemical modications of high-pressure-treated soybean
protein isolates. Journal of Agricultural and Food Chemistry, 52, 1564e1571.
Rackis, J. J., Wolf, W. J., & Baker, E. C. (1986). Protease inhibitors in plant foods:
content and inactivation. In M. Friedman (Ed.), Nutritional and toxicological
signicance of enzyme inhibitors in foods. New York: Plenum Pub. Corp.
Renkema, J. M. S., Gruppen, H., & van Vliet, T. (2002). Inuence of pH and ionic
strength on heat-induced formation and rheological properties of soy protein
gels in relation to denaturation and their protein compositions. Journal of
Agricultural and Food Chemistry, 50, 6064e6071.
Renkema, J. M. S., & van Vliet, T. (2004). Concentration dependence of dynamic
moduli of heat-induced soy protein gels. Food Hydrocolloids, 18, 483e487.
Ren, C. G., Tang, L., Zhang, M., & Guo, S. T. (2009a). Interactions between whey
soybean protein (WSP) and beta-conglycinin (7S) during the formation of
protein particles at elevated temperatures. Food Hydrocolloids, 23, 936e941.
Ren, C. G., Tang, L., Zhang, M., & Guo, S. T. (2009b). Structural characterization of
heat-induced protein particles in soy milk. Journal of Agricultural and Food
Chemistry, 57, 1921e1926.
Ringgenberg, E., Alexander, M., & Corredig, M. (2013). Effect of concentration and
incubation temperature on the acid induced aggregation of soymilk. Food Hy-
drocolloids, 30, 463e469.
Rivas, H. J., & Sherman, P. (1984). Soy and meat proteins as emulsion stabilizers. 4.
The stability and interfacial rheology of O/W emulsions stabilised by soy and
meat protein fractions. Colloids and Surfaces, 11, 155e171.
Saio, K., Kamiya, M., & Watanabe, T. (1969). Food processing characteristics of
soybean 11S and 7S proteins. Part I. Effect of difference of protein components
among soybean varieties on formation of tofu-gel. Agricultural and Biological
Chemistry, 33, 1301.
Samoto, M., Maebuchi, M., Miyazaki, C., Kugitani, H., Kohno, M., Hirotsuka, M., et al.
(2007). Abundant proteins associated with lecithin in soy protein isolate. Food
Chemistry, 102, 317e322.
Smith, A. K., Rackis, J. J., Isnardi, P., Cartter, J. L., & Krober, O. A. (1966). Nitrogen
solubility index isolated protein yield and whey nitrogen content of several
soybean strains. Cereal Chemistry, 43, 261e270.
Sorgentini, D. A., & Wagner, J. R. (1999). Comparative study of structural charac-
teristics and thermal behavior of whey and isolate soybean proteins. Journal of
Food Biochemistry, 23, 489e507.
Speroni, F., & Anon, M. C. (2013). Cold-set gelation of high pressure-treated soybean
proteins. Food Hydrocolloids, 33, 85e91.
Tang, C. H., & Liu, F. (2013). Cold, gel-like soy protein emulsions by micro-
uidization: emulsion characteristics, rheological and microstructural proper-
ties, and gelling mechanism. Food Hydrocolloids, 30, 61e72.
318 K. Nishinari et al. / Food Hydrocolloids 39 (2014) 301e318
Tang, C. H., Luo, L. J., Liu, F., & Chen, Z. (2013). Transglutaminase-set soy globulin-
stabilized emulsion gels: Inuence of soy beta-conglycinin/glycinin ratio on
properties, microstructure and gelling mechanism. Food Research International,
51, 804e812.
Tang, C. H., & Wang, C. S. (2010). Formation and characterization of amyloid-like
brils from soy beta-conglycinin and glycinin. Journal of Agricultural and Food
Chemistry, 58, 11058e11066.
Tang, C. H., Wu, H., Chen, Z., & Yang, X. Q. (2006). Formation and properties of
glycinin-rich and beta-conglycinin-rich soy protein isolate gels induced by
microbial transglutaminase. Food Research International, 39, 87e97.
Tran, T., & Rousseau, D. (2013). Stabilization of acidic soy protein-based
dispersions and emulsions by soy soluble polysaccharides. Food Hydrocolloids,
30, 382e392.
Tsumura, K., Saito, T., Tsuge, K., Ashida, H., Kugimiya, W., & Inouye, K. (2005).
Functional properties of soy protein hydrolysates obtained by selective prote-
olysis. Lwt-Food Science and Technology, 38, 255e261.
Utsumi, S., Damodaran, S., & Kinsella, J. E. (1984). Heat-induced interactions be-
tween soybean proteins: preferential association of 11S basic subunits and beta
subunits of 7S. Journal of Agricultural and Food Chemistry, 32, 1406e1412.
Utsumi, S., Maruyama, N., Satoh, R., & Adachi, M. (2002). Structure-function re-
lationships of soybean proteins revealed by using recombinant systems. Enzyme
and Microbial Technology, 30, 284e288.
Utsumi, S., Matsumura, Y., & Mori, T. (1997). In S. P. Damodaran (Ed.), Food proteins
and TheirApplications. New York: Dekker. A.
Utsumi, S., Nakamura, T., & Mori, T. (1982). A micro-method for the measurement of
gel properties of soybean 11S globulin. Agricultural and Biological Chemistry, 46,
1923e1924.
Veerman, C., Sagis, L. M. C., & van der Linden, E. (2003). Gels at extremely low
weight fractions (0.07%) formed by irreversible self-assembly of proteins.
Macromolecular Bioscience, 3, 243e247.
Wagner, J. R., & Gueguen, J. (1995). Effects of dissociation, deamidation, and
reducing treatment on structural and surface active properties of soy glycinin.
Journal of Agricultural and Food Chemistry, 43, 1993e2000.
Wagner, J. R., & Gueguen, J. (1999). Surface functional properties of native, acid-
treated, and reduced soy glycinin 2. Emulsifying properties. Journal of Agricul-
tural and Food Chemistry, 47, 2181e2187.
Wang, C. H., & Damodaran, S. (1991). Thermal gelation of globular proteins: inu-
ence of protein conformation on gel strength. Journal of Agricultural and Food
Chemistry, 39, 433e438.
Wan, Z. L., Wang, J. M., Wang, L. Y., Yang, X. Q., & Yuan, Y. (2013). Enhanced physical
and oxidative stabilities of soy protein-based emulsions by incorporation of a
water-soluble stevioside-resveratrol complex. Journal of Agricultural and Food
Chemistry, 61, 4433e4440.
Watanabe, T., Ebine, H., & Ota, T. (1991). Soybean foods. Tokyo: Kohrin.
Wongprecha, T., Takaya, T., Kawase, T., Nagano, T., & Nishinari, K. (2000). In
K. Nishinari (Ed.), Hydrocolloids part 1 physical chemistry and industrial application
of gels, polysaccharides, and proteins (pp. 435e439). Amsterdam: Elsevier Science.
Wu, S. W., Murphy, P. A., Johnson, L. A., Fratzke, A. R., & Reuber, M. A. (1999). Pilot-
plant fractionation of soybean glycinin and beta-conglycinin. Journal of the
American Oil Chemists Society, 76, 285e293.
Wu, N. N., Yang, X. Q., Teng, Z., Yin, S. W., Zhu, J. H., & Qi, J. R. (2011). Stabilization of
soybean oil body emulsions using kappa, iota, lambda-carrageenan at different
pH values. Food Research International, 44, 1059e1068.
Yasir, S. B. M., Sutton, K. H., Newberry, M. P., Andrews, N. R., & Gerrard, J. A. (2007).
The impact of transglutaminase on soy proteins and tofu texture. Food Chem-
istry, 104, 1491e1501.
Yoshida, M., Kohyama, K., & Nishinari, K. (1992). Gelation properties of soymilk and
soybean 11S globulin from Japanese-grown soybeans. Bioscience Biotechnology
and Biochemistry, 56, 725e728.
Yoshimura, M., Naito, S., Nagano, T., & Nishinari, K. (2007). Effect of concentration
of soybean powder on the rheological properties and the network structure
of soybean curd prepared from powdered soybean. Journal of the Japanese Society
for Food Science and Technology-Nippon Shokuhin Kagaku Kogaku Kaishi, 54,143e151.
Yuan, Y., Wan, Z. L., Yin, S. W., Yang, X. Q., Qi, J. R., Liu, G. Q., et al. (2013). Charac-
terization of complexes of soy protein and chitosan heated at low pH. LWT-Food
Science and Technology, 50, 657e664.