Carbohydrate Polymers 134 (2015) 119–127

Contents lists available at ScienceDirect

Carbohydrate Polymers
journal homepage: www.elsevier.com/locate/carbpol

Interactions between soy protein from water-soluble soy extract and

polysaccharides in solutions with polydextrose
Jordana C. Spada ∗ , Ligia D.F. Marczak, Isabel C. Tessaro, Nilo S.M. Cardozo
Chemical Engineering Department, Federal University of Rio Grande do Sul, Rua Engenheiro Luiz Englert s/n, Porto Alegre, RS 90040-040, Brazil

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

Article history: This study focuses on the investigation of the interactions between polysaccharides (carrageenan and
Received 21 February 2015 carboxymethylcellulose – CMC) and soy proteins from the water-soluble soy extract. The influence
Received in revised form 11 July 2015 of pH (2–7) and protein–polysaccharide ratio (5:1–40:1) on the interaction between these polyelec-
Accepted 21 July 2015
trolytes was investigated in aqueous solutions with 10% of polydextrose and without polydextrose.
Available online 26 July 2015
The studied systems were analyzed in terms of pH-solubility profile of protein, ␨-potential, methy-
lene blue–polysaccharide interactions, differential scanning calorimetry (DSC), Fourier transform infrared
Keywords:
spectroscopy (FTIR), and confocal laser scanning microscopy. Although the mixtures of soy extract with
Interactions
Soy protein both carrageenan and CMC showed dependency on the pH and protein–polysaccharide ratio, they did not
Carrageenan present the same behavior. Both polysaccharides modified the pH-solubility profile of the soy protein,
Carboxymethylcellulose shifting the pH range in which the coacervate is formed to a lower pH region with the decrease of the soy
Polydextrose extract–polysaccharide ratio. The samples also presented detectable differences regarding to ␨-potential,
DSC, FTIR and microscopy analyses. The complex formation was also detected even in a pH range where
both biopolymers were net-negatively charged. The changes promoted by the presence of polydextrose
were mainly detected by blue–polysaccharide interactions measures and confocal microscopy.
© 2015 Elsevier Ltd. All rights reserved.

1. Introduction plex coacervation, when protein and polysaccharide attract each

other, forming a two phase system consisting of a concentrated
Polysaccharides and proteins are important classes of food protein/polysaccharide phase and a dilute phase. By changing
components. They may affect structure, stability and texture of ingredients and/or solvent properties, a system may shift from one
food products due to their functional properties, such as gelling, scenario to another (De Kruif, Weinbreck, & De Vries, 2004).
thickening and surface stabilizing effects (Tolstoguzov, 1991). Proteins can attract or repel polysaccharides depending on their
Charge-carrying polysaccharides such as carrageenan and car- origin, the conditions to which they are submitted (medium pH,
boxymethylcellulose (CMC) can interact with proteins forming ionic strength, temperature, concentration, shear, etc.), and also on
soluble or non-soluble complexes capable of modifying the afore- the presence of other components/ligands (Delben & Stefancich,
mentioned properties (Stone, Cheung, Chang, & Nickerson, 2013; 1997). Due to the fact that polysaccharide-protein complexes may
Vikelouda & Kiosseoglou, 2004). For anionic polysaccharides like show different properties from those of each individual biopoly-
these, the interaction between carboxyl or sulfate ions with the mer, they can be used in food applications with different purposes
positively charged surface protein groups is expected to take place such as stabilizing air-water or oil-water interfaces or modifying
at pH values between the pKa of the polysaccharide and the iso- texture (viscosifying and gelling properties).
electric point of the protein (Tolstoguzov, 1997). When charged Since soy protein presents a well-balanced amino-acid composi-
proteins and polysaccharides (also known as polyelectrolytes) are tion, it is considered a high-quality protein, showing great potential
mixed, three different scenarios are possible: (i) segregative phase in the substitution of meat and dairy proteins in comparison to pro-
separation, when the ingredients repel each other and two phases teins from other vegetable sources (Friedman & Brandon, 2001). Soy
are formed; (ii) co-solubility, when the ingredients mix well and derivatives as water-soluble soy extract are widely employed in
the solution is stable; and (iii) associative phase separation or com- food industry due to its high protein concentration (∼44%) (Brasil,
1978), low cost and easy addition to the formulation.
Several recently published research papers have been focused
∗ Corresponding author. on the interaction between charge-carrying polysaccharides and
E-mail address: [email protected] (J.C. Spada). different proteins, including milk proteins (Benichou, Aserin, Lutz,

http://dx.doi.org/10.1016/j.carbpol.2015.07.075
0144-8617/© 2015 Elsevier Ltd. All rights reserved.
120 J.C. Spada et al. / Carbohydrate Polymers 134 (2015) 119–127

& Garti, 2007; Bedie, Turgeon, & Makhlouf, 2008; Ercelebi &
Ibanoglu, 2007), soy proteins (Braga, Azevedo, Julia Marques,
Menossi, & Cunha, 2006; Molina Ortiz Puppo, & Wagner, 2004;
Yuan, Wan, Yang, & Yin, 2014), and gelatin and chitosan (Michon,
Vigouroux, Boulenguer, Cuvelier, & Launay, 2000; Yuan et al., 2014).
Investigations on the effect of the addition of co-solutes like poly-
dextrose on gelling conditions have also been reported since they
are capable of modifying the morphology of three dimensional
structures, leading to the formation of gels with rubbery viscoelas-
ticity (Almrhag, George, Bannikova, Katopo, & Kasapis, 2012). Some
current food formulations contain polydextrose as a bulking agent,
in proportions from 4 to 10 % in low sugar products and from 2 to 5
% for low fat products (Mitchell, 2002). Polydextrose is a randomly
linked polymer of glucose that has virtually no sweetness and an
energy value of only 1 kcal/g; its physiological benefits as dietary
fiber, low glycemic index, and prebiotic effects make it a useful
ingredient in novel food and nutraceutical applications (Mitchell,
1996). In spite of the wide application of polydextrose in food indus-
try, to the best of our knowledge, there is no information about the
influence of the addition of this co-solute in solutions which present
protein–polysaccharide interactions.
In the present work, the influence of pH and
protein–polysaccharide ratio on the interaction between these
polyelectrolytes in aqueous solutions, with polydextrose (10%)
and without polydextrose, was investigated through methylene
blue–polysaccharide interactions, ␨-potential, pH-solubility profile
of protein, differential scanning calorimetry, Fourier transform
infrared spectroscopy, and microscopy analysis.

2. Material and methods Fig. 1. (a) UV spectrum of methylene-blue (MB), carrageenan–MB and CMC–MB. (b)
Curve of the ratio between 664 nm and 557 nm versus carrageenan concentration.
2.1. Materials

and HNO3 . Using the elctrophoretic mobility determined with

The water-soluble soy extract (denoted by SE) was provided by
microelctrophoresis equipment, the Henry equation was applied
Olvebra Industrial S/A (Eldorado do Sul, Brazil). According to the
for calculating the ␨-potential (expressed in mV). All measurements
company information, the physicochemical composition of the soy
were conducted in duplicate.
extract was 10% of moisture, 18.5% of carbohydrates, 44% of pro-
tein (11S and 7S globulins; the ratio 11S/7S ranged between 1.8
and 2), 26% of fat and 1.5% of fiber. The total content of protein 2.4. pH-solubility profile of protein
was confirmed by Kjeldahl analysis. Carrageenan with more than
90% of lambda form (SatiagumTM ADC 25) was provided by Cargill Solutions of different soy extract–polysaccharide ratios were
Alimentos (Minnesota, United States) and carboxymethylcellulose prepared by mixing the appropriate volumes of stock solutions.
was donated by Hexus Food (Portão, Brazil). Litesse® (Danisco Brasil Protein solubility as a function of pH (2–7) was determined for
Ltda., Cotia, Brazil) was the raw material used as source of polydex- water-soluble soy extract (SE) and for soy extract–polysaccharides
trose. The powder was of 90% purity with 4% moisture and with an samples. Water dispersions were magnetically stirred at 120 rpm
average degree of polymerization of 12 glucose residues. for 1 h at room temperature (25 ± 1 ◦ C) and pH was adjusted to the
desired value with diluted solutions of HCl or NaOH. Dispersions
2.2. Sample preparation were centrifuged at 10,000 × g for 10 min at 10 ◦ C, and protein con-
tent in the supernatant was measured by Lowry method (Lowry,
Water-soluble soy extract (5%) and polysaccharides (0.1%) solu- Rosenbroug, Lewis, & Randall, 1951). Solubility was expressed as
tions were prepared with distilled purified water and kept under a percentage (g of soluble protein/0.44 g of isolate in sample). All
refrigeration for 48 h until complete dissolution of the components. solubility determinations were performed in triplicate.
Different soy extract–polysaccharyde ratios (5:1, 10:1, 20:1 and
40:1) were prepared by mixing the appropriate volumes of stock 2.5. Methylene blue–carrageenan interaction measurements
solutions.
According to Benichou et al. (2007) and Michon, Konaté,
2.3. ␨-Potential Cuvelier, and Launay (2002), an alternative method to investi-
gate possible interactions between polysaccharide and protein
The ␨-potentials of the samples were determined using molecules is the analysis of methylene blue–polysaccharide inter-
a microelectrophoresis equipment (Brookhaven Instruments actions through UV-spectroscopy. However, this type of measure
Corporation® , Zeta Plus, NY, USA). Aqueous solutions samples was conducted only for samples with carrageenan since the sam-
(0.01% of solids) with different soy extract–polysaccharide ratios ples with CMC presented the same UV spectrum as blue-methylene
and pH were prepared under magnetic stirring (120 rpm) and (Fig. 1a).
filled into a disposable ␨-potential folded capillary cell (DTS1060). Aqueous solutions of carrageenan and methylene blue (MB)
The mixtures were prepared with aqueous solution of KNO3 with a concentration of 0.1% w/v and 0.001% w/v, respectively,
10−3 mol L−1 and pH was adjusted by the addition of KOH were prepared and then were properly mixed to obtain a series
 J.C. Spada et al. / Carbohydrate Polymers 134 (2015) 119–127 121

of mixed polysaccharide–dye solutions containing 1 mL of 0.001%

w/v MB and 1 mL of carrageenan varying in concentration from
0 w/v to 0.01% w/v. The change in the ratio between the absorbance
obtained at 664 and 557 nm with the increase in polysaccharide
concentration in the mixed dye–carrageenan solutions was then
followed in order to determine whether MB–carrageenan interac-
tion took place or not. In a second experiment, 1 mL of the mixtures
containing 0.002% w/v of carrageenan and different contents of
soy extract at different pHs (2, 5 and 7) were mixed with the dye
and the evolution of the ratio of the absorbance values at 664 nm
and 557 nm was followed as a function of soy extract concentra-
tion as an attempt to search for possible interaction between the
polysaccharide and the protein molecules (Benichou et al., 2007).

2.6. Differential scanning calorimetry (DSC)

To perform the analyses of DSC and FTIR, the stock solutions

of soy extract and polysaccharides, as well as their mixtures, were
freeze-dried (LS 6000, Terroni Equipamentos® , São Carlos, Brazil).
The analyzed samples were 5:1 (pH 2) and 20:1 (pH 2, pH 5, and
pH 7) soy extract–carrageenan mixtures and 5:1 (pH 3, pH 5) and
20:1 (pH 3, pH 5, and pH 7) soy extract–CMC mixtures. This smaller
subset of samples was chosen because it is representative of all
differences and similarities among the studied samples. This fact
was verified based on the results of pH-solubility and ␨-potential
curves. The samples (∼5.0 mg) were accurately weighed into alu-
minum pans, which were hermetically sealed and equilibrated at
room temperature for 2 h before analysis. A sealed empty pan was
used as reference. All DSC scans were obtained using DSC 6000
(PerkinElmer), at 10 ◦ C/min from 20 ◦ C to 250 ◦ C, under nitrogen
and argon atmosphere. The melting temperature (Tm ) was taken
as the temperature corresponding to the maximum of the melting
peak. All experiments were conducted in triplicate. Fig. 2. ␨-Potential of the soy extract together with carrageenan (a) and CMC (b) in
different ratios.

2.7. Fourier transform infrared spectroscopy (FTIR)

Tulsa, OK, USA). Post hoc multiple comparisons were performed by
FTIR analyses were performed on a spectrometer equipped Tukey’s test with the level of confidence of 95%.
with a ZnSe single reflection ATR crystal (Bruker Alpha, Eco-ATR,
USA). The freeze dried mixtures were examined to identify the 3. Results and discussion
nature of molecular interactions between the constituents. Spec-
tra were obtained in absorbance mode for the wavelength range 3.1. ␨-Potential
of 600–4000 cm−1 with a resolution of 4 cm−1 and measurements
were averaged on 16 scans. All spectra were smoothed using the Fig. 2 shows the data of ␨-potential as a function of pH for the soy
Savitzky–Golay function and the areas of the peaks were calculated extract–carrageenan (Fig. 2a) and soy extract-CMC (Fig. 2b) mix-
using the Matlab software (v. 5.3, Mathworks Inc., Natick, USA). tures and for the pure components. The points are joined by lines
just for a better visualization. As can be seen, the ␨-potential of pro-
tein from soy extract attained to zero (isoelectric point) between pH
2.8. Confocal laser scanning microscope (CLSM)
4–5, i.e., in this range the electrostatic repulsive force is reduced.
Similar results were reported by Yuan et al. (2014) who studied
The CLSM-images of the mixtures were recorded at room tem-
the fractions of soy protein 7S and 11S and found their isoelectric
perature using a laser scanning confocal microscope (Olympus
points at pH 4.3 and 5.1, respectively, and by Lam, Shen, Paulsen,
America Inc., Melville, NY), equipped with an inverted microscope
and Corredig (2007) who found the isoelectric point of commercial
(Olympus FV 1000), and a 60× oil-immersion objective. Before
soy protein isolate at pH 4.4. As can be seen in Fig. 2, both polysac-
imaging, the mixtures (protein content of 5 mg/mL) were mixed
charides showed a negative charge that decreases with decreasing
with Rhodamine B solution (8 ␮L of 0.2 wt.% solution per mL sam-
pH over the whole pH range used in the present work, while the
ple) according to Yuan et al. (2014). After that, 20 ␮L of the stained
soy extract presented a positive charge in the range of pH between
mixture was placed on a slide. The incident light was emitted by
2 and 4, and a negative charge for pH > 5. So, it is expected the
a laser beam at 473 nm. Multiple fields of view were examined
strongest electrostatic complexes to form when the charges on the
for each sample and a representative image selected. The stains
two polymers are opposite to one another.
selected showed the protein as green against a black background.
The addition of carrageenan or CMC significantly reduced the
␨-potential (p ≤ 0.05) of protein for, practically, all pH values and
2.9. Statistical analysis protein–polysaccharide ratios. The ␨-potential–pH curves of the
mixtures were intermediates to those of the two pure polymers. As
Data were subjected to one-way analysis of variance (ANOVA) can be seen, for the soy extract–polysaccharide mixtures the point
using the software Statistica for Windows, version 10 (Statsoft Inc., of zero of ␨-potential shifted to lower pH values. The mixtures with
122 J.C. Spada et al. / Carbohydrate Polymers 134 (2015) 119–127

would not have taken place, the polysaccharides would not be

linked to the protein and the soy extract–polysaccharide curves
would have the same profile as the soy extract. Taking into account
the net charge of both molecules (␨-potential results presented in
Fig. 2), it is expected that the complex formation through electro-
static unions predominates in the acidic range, in which the protein
and the polysaccharides have opposite net charges.
Reduction of the protein solubility was also observed for the
mixtures with carrageenan at alkaline pH (6 and 7), although elec-
trostatic interactions between the polymers are not expected in this
pH range, where both polymers have a net negative charge. In such
case, this effect could be attributed to the formation of hydropho-
bic interactions (Molina Ortiz et al., 2004). However, the results
regarding to blue-methylene measurements show that, even at pH
7, the carrageenan has been consumed (results better commented
in Section 3.3), suggesting the existence of electrostatic interactions
in the alkaline region. The occurrence of electrostatic interactions
between anionic polysaccharides and globular proteins at pH val-
ues much above the protein isoelectric point has also been reported
for a number of other protein/gum systems; these electrostatic
interactions were attributed to the heterogeneity of the charge dis-
tribution on the protein surface that may favor specific attractive
interactions between the polysaccharide anionic groups and pos-
itively charged regions of the protein surface, even at a pH where
both biopolymers carry a net negative charge (De Kruif et al., 2004).
In the case of soy extract–CMC mixtures, the solubility of the soy
protein in the alkaline region was practically not affected by the
addition of the polysaccharides.
Fig. 3. Solubility profiles as a function of pH of the soy extract mixtures with car-
rageenan (a) and CMC (b) at different ratios.
In the pH range 4–5, the addition of both polysaccharides
increased the solubility of the protein. This occurs because in the
case of the pure soy extract the soy proteins particles are practically
protein–polysaccharide ratios of 10:1 and 20:1 presented isoelec- uncharged in this pH range, with a predominant tendency to aggre-
tric point between pH 3–4, while both 5:1 mixtures did not present gate. When the polysaccharides (negatively charged) are added to
isoelectric point, showing similar behavior to the pure polysac- the system, a new repulsion force between particles arises from the
charides. It was verified that the presence of polydextrose did not surplus negative charge. This mechanism is the basis of the use of
cause significant changes in the ␨-potential values of both types of others hydrocolloids to prevent protein aggregation and stabilize
mixtures. food systems.

3.2. pH-solubility profile of protein 3.3. Methylene blue–carrageenan interaction experiments

Fig. 3 shows the protein solubility profiles as a function of pH As presented in Fig. 1b, the ratio between the UV absorbances
for the water-soluble soy extract and the soy extract–carrageenan at 664 nm and 557 nm (hereinafter called Abs664/Abs557 ratio) in
(Fig. 3a) and soy extract–CMC (Fig. 3b) mixtures. Again, the exper- the UV spectrum of a dilute methylene blue solution decreases with
imental points are joined for a better visualization. The protein the addition of carrageenan. This decrease has been attributed to
solubility of the samples with polydextrose was not measured an electrostatic interaction between the polysaccharide and the
because sugars or polyols interfere in the Lowry methodology. dye molecules (Michon et al., 2002). Also according to Fig. 1b, all
The soy extract sample and the 40:1 mixtures with both polysac- solutions with carrageenan concentrations higher than 0.002% pre-
charides showed a typical U-shaped pH–solubility profile, with a sented practically the same Abs664/Abs557 ratio, possibly because
minimum of solubility at about pH 4.5 and 4.0, respectively. The all dye interacted with the polysaccharide. So, the study of the soy
solubility profile of the soy extract sample was similar, but with extract–carrageenan mixtures, the carrageenan concentration was
some different percentage values, to those reported by Malhotra set the concentration at 0.002%.
and Coupland (2004) and Molina Ortiz et al. (2004), who studied Table 1 shows the values of Abs664/Abs557 ratio for the soy
isolate soy protein. In this sense it is important to point out that extract–carrageenan samples with and without polydextrose at pH
the solubility of proteins is related to contents of the four main 2, 5 and 7. The addition of soy protein to the polysaccharide–dye
fractioned proteins (2S, 7S, 11S, and 15S, classified according to the solution increases the Abs664/Abs557 ratio which, according to
sedimentation velocity) which, in turn, depend on the character- Benichou et al. (2007), should be attributed to the weakening of
istics of the industrial process used for its extraction or isolation. the polysaccharide–methylene blue interaction as a result of the
This is probably the cause for the aforementioned differences in the protein binding to the polysaccharide molecules. When compared
solubility profiles. to the samples containing only carrageenan (ratio 0), all mixtures at
When the polysaccharides were added, lower solubility values pH 2, and 20:1 and 40:1 mixtures at pHs 5 and 7 showed significant
were observed at lower pH. This result can be explained by consid- higher Abs664/Abs557 values.
ering that in specific conditions (pH and polymer concentration) From Table 1 is also possible to observe that the presence of
soy proteins tend to form insoluble and electrically neutral com- polydextrose also led to changes in the soy extract–carrageenan
plex with polysaccharides molecules; the aggregation of insoluble interactions. A possible interaction between carrageenan and poly-
complex particles is mainly due to electrostatic and hydrophobic dextrose was discarded, since the ratio Abs664/Abs557 values were
interactions (Tolstoguzov, 2003). If the formation of the complex statically equal in the absence or the presence of polydextrose
 J.C. Spada et al. / Carbohydrate Polymers 134 (2015) 119–127 123

Table 1 tively. The peak at 1587 cm−1 was assigned to the carboxylate
Results of methylene-blue measurements of the soy extract–carrageenan samples
(COO− ) group (Zaleska, Tomasik, & Lii, 2002).
with and without polydextrose.
As well as for CMC, the FTIR spectrum of carrageenan (Fig. 4a)
Samples Abs664/Abs557 ratio also showed a large band at around 3200–3400 cm−1 related
pH Ratio Without polydextrose* With 10% polydextrose* to O H stretching vibrations and C H stretching vibrations at
2930 cm−1 . The band corresponding to the carboxylic group was
2 0 0.98 ± 0.06 Ea
1.01 ± 0.02 Da
5 2.32 ± 0.09Da 2.04 ± 0.04Ca detected at around 1560–1780 cm−1 and the band at around
10 8.92 ± 0.13Aa 8.56 ± 0.54Aa 1220–1260 cm−1 was related to the total sulphate content. The
20 8.81 ± 0.44Aa 9.28 ± 0.83Aa spectrum presented two bands, at 830 and 820 cm−1 , corre-
40 8.35 ± 0.43Aa 7.76 ± 0.73Aa
sponding to galactose-2-sulphate and to galactose-6-sulphate,
5 0 0.59 ± 0.03Ha 0.62 ± 0.01 Ha respectively, which indicate the characteristic absorption pattern
5 0.61 ± 0.05Hb 0.76 ± 0.02Fa of lambda carrageenan (Roberts & Quemene, 1999; Stephanie, Eric,
10 0.68 ± 0.01Hb 0.88 ± 0.03Da
Sophie, Christian, & Yu, 2010). Moreover, peaks at 930 cm−1 (3,6-
20 3.10 ± 0.17Cb 4.93 ± 0.25Ba
40 5.43 ± 0.15Bb 8.78 ± 0.65Aa anhydro-d-galactose residue), 805 cm−1 (3,6-anhydrogalactose-
2-sulphate) and 850–840 cm−1 (axial sulfate ester at O-4 of
7 0 0.63 ± 0.02Ha 0.57 ± 0.04 Ha
5 0.58 ± 0.02Hb 0.67 ± 0.02Ga
the 3-linked galactose residue) have been identified, indicat-
10 0.67 ± 0.08Hb 0.76 ± 0.01Fa ing the presence of Ãota carrageenan. From Fig. 4c, it can be
20 0.74 ± 0.03Gb 0.85 ± 0.03Ea seen that the specific chemical bonds within the polydextrose
40 0.84 ± 0.02Fa 0.88 ± 0.02Ea molecule were detected: O H stretching (3500 cm−1 ), C H stretch-
*Means followed by different superscripts in capital letter are significantly differ- ing (2900 cm−1 ), C O stretching of aldehyde (1627 cm−1 ) and
ent at p < 0.05 using Tukey’s test in the same column. Means followed by different stretching vibration of C O C glycosidic linkage (1180–930 cm−1 )
superscripts in lowercase letter are significantly different at p < 0.05 using Tukey’s (Mickova, Copikova, & Synytsya, 2007).
test in the same line.
For the soy extract spectrum, the peak in the range
3300–3400 cm−1 probably results from the overlapping contri-
(samples named as ratio 0). Based on these results and on the butions of the stretching vibration of both the O H groups
information given by Miyawaki (2007) who reported that protein and N H groups. The absorption bands at 1630–1710 cm−1 and
stability can be affected by a coexisting solute as the third com- 1520–1590 cm−1 are assigned, respectively, to the amide I and
ponent, it is suggested that the protein group has been affected by amide II groups of the soy protein (Schmidt, Giacomelli, & Soldi,
the presence of polydextrose. In accordance with Miyawaki (2007), 2005; Wang et al., 2014). The amide II is characteristic of the com-
proteins are stabilized by solutes that are preferentially-excluded bined movement of the N H vibrations and C N stretching from
from their surface, while preferentially-bound solutes destabilize the CO NH group (Bandekar, 1992; Mangavel, Barbot, Popineau,
proteins. Proteins with more unfolded conformation (destabilized) & Guéguen, 2001). The amide I band is mainly attributed to C O
expose more reactive sites (amino acids) to the solvent phase, so stretching of the peptide bond and is sensitive to the secondary
interactions with the polysaccharide would be more likely to occur. structure of protein backbone. Consequently, it is affected by con-
Based on this and on the results from Table 1, it is believed that the formational changes in the protein molecules, being the most
polydextrose has caused an increased in the destabilization of the commonly used band for the quantitative analysis of secondary
protein. structures. Therefore, it is expected that this band may be used for
The differences between the samples with and without poly- assessing the influence of pH and protein–polysaccharides ratio on
dextrose were significantly different for all ratios at pH 5, and this the structural and interactional features of the studied mixtures.
result is consistent with the information reported by Damodaran,
Parkin and Fennema (2008). These authors have reported that pro-
teins are more stable (folded state) at their isoelectric point than 3.4.2. FTIR of protein–polysaccharides mixtures
at any other pH, while at extreme pH values, strong intramolecular From Fig. 4 it can also be observed that the positions of the
electrostatic repulsion caused by high net charge results in swelling bands assigned to the amide I and II regions remained constant
and unfolding of the protein molecule. The more pronounced effect in the mixtures. However, the band assigned to carboxylate in
of the addition of the polydextrose on the soy extract–carrageenan the polysaccharides at 1587 cm−1 for CMC was not detectable in
interaction at pH 5 can be related to possible unfolding of the pro- the mixtures with soy extract, indicating interactions between the
teins, which are practically folded in this pH. Differently, at pH 2 the protein and the polysaccharide. Based on the aspects discussed in
proteins are already unfolded; and at pH 7 the proteins and polysac- the previous section, the changes resulting from the interactions
charides are both negatively charged preventing the extension of between protein and polysaccharides were quantitatively mea-
the interaction between them. sured by calculating the normalized area of the peak related to
amide I (1628 cm−1 ), i.e., its area divided by that of the C H peak
(2930 cm−1 ), used as internal reference. The well-defined C H peak
3.4. Fourier transform infrared spectroscopy (FTIR) at 2930 cm−1 due represents a good internal reference for the total
carbohydrate content, because the number of C H groups remains
This section deals with molecular aspects of the constituents and constant, irrespective of changes in the protein–polysaccharides
the nature of their molecular interactions using FTIR. Fig. 4(a–c) ratio (Rochas, Lahaye, & Yaphe, 1896).
reproduces the FTIR spectra of individual components (car- Table 2 shows the values of normalized area of the amide I peak
rageenan, CMC, soy extract and polydextrose) and of their mixtures. for the different samples. As can be seen, these values decreased
For purposes of clarity, the results obtained for pure components when the pH decreased (from 7 to 2) and also when the ratio
and mixtures are discussed separately below. between the polymers decreased (from 20:1 to 5:1). The changes
were higher for mixtures containing carrageenan. These results can
3.4.1. FTIR of individual components be taken as an indicative of the degree of interactions occurring
FTIR spectrum of CMC (Fig. 4b) showed broad absorption between proteins and polysaccharides. Due to the higher propor-
bands at 3200–3400 cm−1 , 2800–3000 cm−1 and 1100–1300 cm−1 , tion of the polydextrose in relation to the other components, the
attributed to O H, C H, and C O C stretching vibration, respec- FTIR of the mixtures containing this co-solute were very similar to
124 J.C. Spada et al. / Carbohydrate Polymers 134 (2015) 119–127

Fig. 4. FTIR absorbance spectra of the lyophilized samples containing soy extract and carrageenan (a), soy extract and CMC (b) and these components mixed with polydextrose
(c).

Table 2 CLSM-images of Fig. 5 are supplied as additional material, for better

Normalized areas of the amide I FTIR bands (1628 cm−1 ) for different samples.
visualization.
Samples Condition Normalized area As can be seen, the soy extract (control) showed aggregates at
Amid I/CH pH close to its isoelectric point (Fig. 5, SE pH5). Both mixtures
1628/2930 cm−1
(with carrageenan or CMC) showed more homogenized struc-
Control Pure soy extract 0.773 ture and smaller size and number of aggregates compared to the
Soy 20:1 pH 3 0.541 control sample at that pH. At lower pHs, the mixtures showed
extract–CMC 20:1 pH 5 0.625 dense and large coacervates, indicating phase separation. Taking
20:1 pH 7 0.761 into consideration the ␨-potential results discussed in Section 3.1,
5:1 pH 3 0.426
this behavior can be attributed to the strong interaction between
5:1 pH 5 0.527
the two oppositely charged biopolymers. This interaction induces
Soy 20:1 pH 2 0.490 charge neutralization in the mixtures, causing agglomeration and
extract–carrageenan 20:1 pH 5 0.588
increase of the polymer particles size. It can also be clearly observed
20:1 pH 7 0.680
5:1 pH 2 0.409 from Fig. 5 that primary complexes are initially formed (Fig. 5, 20:1
pH 5) and, then, coarsened when pH was reduced (Fig. 5, 20:1 pH
2), suggesting a nucleation and a growth mechanism. Furthermore,
that of the pure polydextrose (Fig. 4c). So, it was not possible to soluble complexes were formed in the mixtures at higher pH. The
quantify the normalized areas for these samples. CLSM-images also suggest that the microstructure of the mixtures
was dependent on the kind of the polysaccharide and the soy pro-
3.5. Microstructure analysis tein/polysaccharide ratio. In the presence of polysaccharides, the
particles of soy extract were more distant from each other. The
Fig. 5 presents CLSM-images of the mixtures as a function of intrinsic viscosity values of the polysaccharides are different and
pH and soy extract/polysaccharide ratio. Magnified versions of the
 J.C. Spada et al. / Carbohydrate Polymers 134 (2015) 119–127 125

Fig. 5. Confocal laser scanning photomicrographs of the soy extract (SE) and the mixtures containing soy extract–carrageenan and soy extract–CMC as a function of ratio and
pH (protein content = 5 mg/mL). Green color representes the protein phase and the polysaccharides are not stained. The white scale bar represents 100 ␮m. (For interpretation
of the references to color in this figure legend, the reader is referred to the web version of this article.)
126 J.C. Spada et al. / Carbohydrate Polymers 134 (2015) 119–127

Fig. 6. DSC heating thermograms of the lyophilized mixtures containing soy extract–carrageenan (a) and soy extract–CMC (b) at different pH and ratios and in the presence
of polydextrose.

this fact can have affected the observed microstructure, preventing The polydextrose shows a broad endothermic peak between
the approach of particles and aggregates in the cover slip. 125 ◦ C and 160 ◦ C. Similarly, Ribeiro, Zimeri, Yildiz, and Kokini
At lower pH values, the pure soy extract particles seem more (2003) analyzing the thermograms of the polydextrose at differ-
swollen. As discussed previously, at extreme pH values, strong ent values of water activity (aw ), found melting endotherm ranging
intramolecular electrostatic repulsion caused by high net charge approximately from 130 to 150 ◦ C for the samples with low aw
results in swelling and unfolding of the protein molecule. The pres- (0.08 and 0.37). The thermograms of the mixtures with polydex-
ence of polydextrose also cause a slight swelling in the particles, trose were similar to those of the pure polydextrose, probably due
which corroborates the discussion in Section 3.3. to the large quantity of this component. Similar behavior was found
in the FTIR analysis, as commented in Section 3.4.

3.6. Thermal behavior 4. Conclusion

Fig. 6 shows the heating thermograms of the lyophilized sam- The results showed that the physicochemical behavior of the
ples. The thermogram of the carrageenan powder (Fig. 6a) shows soy extract–polysaccharide mixtures at different pH (2–7) and
two endothermic peaks, at 165 and 180 ◦ C, and also an exothermic soy extract/polysaccharide ratios (5:1–40:1) differ from that of
transition at about 150 ◦ C. In accordance with Brown (2001), as the the pure soy extract, mainly in terms of solubility, charge and
temperature is slowly increased, organic polymers may recrystal- phases morphology. The FTIR and DSC results also indicated the
lize giving an exothermic peak, before melting point. formation of considerable interactions between the studied com-
The pure soy extract presented two endothermic peaks, at 140 ponents. Moreover, the polydextrose showed interactions with the
and 175 ◦ C, that can be related to the fractions of soy protein, 7S and soy extract which were detected by blue-methylene interactions
11S, respectively; and practically all the soy extract–polysaccharide measurements and confocal microscopy. Overall, this work pro-
mixtures presented only one broad endothermic peak, except the vides fundamental information about the interactions between the
soy extract–carrageenan mixture ratio of 5:1 at pH 2 that also polysaccharides (carrageenan and CMC) and soy extract in aqueous
showed an exothermic peak. Su, Huang, Yuan, Wang, and Li (2010) solutions with different pHs and in the presence of polydextrose.
studied blends of carboxymethylcellulose–soy protein isolate and The results indicated that the soy extract, although to be not an iso-
found a melting temperature between 140 ◦ C and 160 ◦ C, which is late source of protein, showed interactions with the carrageenan,
similar to the values for the the soy extract–CMC mixtures in the CMC and polydextrose. These interactions suggest the best condi-
present work. tions under which such mixtures can be used in food formulations
It is worthwhile to observe that both position and the shape (creams, sauces, yogurts, etc.) providing greater stability.
of the melting peak of the mixtures were remarkably dependent
on the pH and soy protein/polysaccharide ratio. In the samples Appendix A. Supplementary data
for which the complex formation was confirmed by the confocal
microscopy (20:1 soy extract–carrageenan at pH 2 and 20:1 soy Supplementary data associated with this article can be found, in
extract–CMC at pH 3), the melting peak was narrower than for the online version, at http://dx.doi.org/10.1016/j.carbpol.2015.07.
the other mixtures analyzed. When the complex formation did not 075
take place, the mixtures showed broad peaks, which may indicate
that the more heterogeneous nature of the mixtures under these References
conditions leads to imperfections in the crystalline structure. For
example, the soy extract–CMC 20:1 at pH 7 showed a peak between Almrhag, O., George, P., Bannikova, A., Katopo, L., & Kasapis, S. (2012). Networks of
polysaccharides with hydrophilic and hydrophobic characteristics in the
102 and 184 ◦ C, unlike the sample at pH 3 that showed a narrow presence of co-solute. International Journal of Biological Macromolecules, 51,
peak between 183 and 195 ◦ C. 138–145.
 J.C. Spada et al. / Carbohydrate Polymers 134 (2015) 119–127 127

Bandekar, J. (1992). Amide modes and protein conformation. Biochimica et Miyawaki, O. (2007). Hydration state change of proteins upon unfolding in sugar
Biophysica Acta – Protein Structure and Molecular Enzymology, 1120, 123–143. solutions. Biochimica et Biophysica Acta, 1774, 928–935.
Bedie, G. K., Turgeon, S. L., & Makhlouf, J. (2008). Formation of native whey protein Molina Ortiz, S. E., Puppo, M. C., & Wagner, J. R. (2004). Relationship between
isolate–low methoxyl pectin complexes as a matrix for hydro-soluble food structural changes and functional properties of soy protein
ingredient entrapment in acidic foods. Food Hydrocolloids, 22, 836–844. isolates–carrageenan systems. Food Hydrocolloids, 18, 1045–1053.
Benichou, A., Aserin, A., Lutz, R., & Garti, N. (2007). Formation and characterization Ribeiro, C., Zimeri, J. E., Yildiz, E., & Kokini, J. L. (2003). Estimation of effective
of amphiphilic conjugates of whey protein isolate (WPI)/xanthan to improve diffusivities and glass transition temperature of polydextrose as a function of
surface activity. Food Hydrocolloids, 21, 379–391. moisture content. Carbohydrate Polymers, 51, 273–280.
Braga, A. L. M., Azevedo, A., Julia Marques, M., Menossi, M., & Cunha, R. L. (2006). Roberts, M. A., & Quemene, B. (1999). Measurement of carrageenans in food:
Interactions between soy protein isolate and xanthan in heat-induced gels: the challenges, progress, and trends in analysis. Trends in Food Science &
effect of salt addition. Food Hydrocolloids, 20, 1178–1189. Technology, 10, 169–181.
Brasil (1978) Ministério da saúde. Agência nacional de vigilância sanitária. Rochas, C., Lahaye, M., & Yaphe, W. (1986). Sulfate content of carrageenan and
Resolução CNNPA n◦ 14, de 28 de junho de 1978. Disponível em: http://www. agardetermined by infrared spectroscopy. Botanica Marina, 29, 335–340.
anvisa.gov.br/anvisalegis/resol/14 78.htm Schmidt, V., Giacomelli, C., & Soldi, V. (2005). Thermal stability of films formed by
Brown, M. E. (2001). Introduction to thermal analysis: techniques and applications. soy protein isolate–sodium dodecyl sulfate. Polymer Degradation and Stability,
Dordrecht: Kluwer Academic Publishers. 87, 25–31.
Damodaran, S., Parkin, K. L., & Fennema, O. R. (2008). . Fennema’s food chemistry Stephanie, B., Eric, D., Sophie, F. M., Christian, B., & Yu, G. (2010). Carrageenan from
(vol. 4) Boca Raton: CRC Press/Taylor & Francis. Solieria chordalis (Gigartinales): structural analysis and immunological
De Kruif, C. G., Weinbreck, F., & De Vries, R. (2004). Complex coacervation of activities of the low molecular weight fractions. Carbohydrate Polymers, 81,
proteins and anionic polysaccharides. Current Opinion in Colloid & Interface 448–460.
Science, 9, 340–349. Stone, A. K., Cheung, L., Chang, C., & Nickerson, M. T. (2013). Formation and
Delben, F., & Stefancich, S. I. (1997). Interaction of food proteins with functionality of soluble and insoluble electrostatic complexes within mixtures
Polysaccharides I: properties upon mixing. Journal of Food Engineering, 31, of canola protein isolate and (␬-, ␫- and ␭-type) carrageenan. Food Research
325–346. International, 54, 195–202.
Ercelebi, E. A., & Ibanoglu, E. (2007). Influence of hydrocolloids on phase separation Su, J.-F., Huang, Z., Yuan, X.-Y., Wang, X.-Y., & Li, M. (2010). Structure and
and emulsion properties of whey protein isolate. Journal of Food Engineering, properties of carboxymethyl cellulose/soy protein isolate blend edible films
80, 454–459. crosslinked by Maillard reactions. Carbohydrate Polymers, 79, 145–153.
Friedman, M., & Brandon, D. L. (2001). Nutritional and health benefits of soy Tolstoguzov, V. (2003). Some thermodynamic considerations in food formulation.
proteins. Journal of Agricultural and Food Chemistry, 49, 1069–1086. Food Hydrocolloids, 17, 1–23.
Lam, M., Shen, R., Paulsen, P., & Corredig, M. (2007). Pectin stabilization of soy Tolstoguzov, V. B. (1991). Functional properties of food proteins and role of
protein isolates at low pH. Food Research International, 40, 101–110. protein–polysaccharide interactions review. Food Hydrocolloids, 4, 429–468.
Lowry, O. H., Rosenbroug, H. J., Lewis, A., & Randall, K. J. (1951). Measurements Tolstoguzov, V. B. (1997). Protein–polysaccharide interactions. In S. Damodaran, &
with the folin–phenol reagents. Journal of Biological Chemistry, 193. A. Paraf (Eds.), Food proteins and their applications (pp. 171–256). New York:
Malhotra, A., & Coupland, J. N. (2004). The effect of surfactants on the solubility, Marcel Dekker.
zeta potential, and viscosity of soy protein isolates. Food Hydrocolloids, 18, Vikelouda, M., & Kiosseoglou, V. (2004). The use of carboxymethylcellulose to
101–108. recover potato proteins and control their functional properties. Food
Mangavel, C., Barbot, J., Popineau, Y., & Guéguen, J. (2001). Evolution of wheat Hydrocolloids, 18, 21–27.
gliadins conformation during film formation: a Fourier transform infrared Wang, L., Xiao, M., Dai, S., Song, J., Ni, X., Fang, Y., et al. (2014). Interactions
study. Journal of Agricultural and Food Chemistry, 49, 867–872. between carboxymethyl konjac glucomannan and soy protein isolate in
Michon, C., Konaté, K., Cuvelier, G., & Launay, B. (2002). Gelatin/carrageenan blended films. Carbohydrate Polymers, 101, 136–145.
interactions in coil and ordered conformations followed by a methylene blue Yuan, Y., Wan, Z.-L., Yang, X.-Q., & Yin, S.-W. (2014). Associative interactions
spectrophotometric method. Food Hydrocolloids, 16, 613–618. between chitosan and soy protein fractions: effects of pH, mixing ratio, heat
Michon, C., Vigouroux, F., Boulenguer, P., Cuvelier, G., & Launay, B. (2000). treatment and ionic strength. Food Research International, 55, 207–214.
Gelatin/iota–carrageenan interactions in non-gelling conditions. Food Zaleska, H., Tomasik, P., & Lii, C. Y. (2002). Formation of carboxymethyl
Hydrocolloids, 14, 203–208. cellulose–casein complexes by electrosynthesis. Food Hydrocolloids, 16,
Mickova, K., Copikova, J., & Synytsya, A. (2007). Determination of polydextrose as a 215–224.
fat replacer in butter. Czech Journal of Food Science, 25, 25–31.
Mitchell, H. (2002). O uso de carboidratos especiais. Food Ingredients, 42–44.
Mitchell, H. L. (1996). The role of the bulking agent polydextrose in fat
replacement. In S. Roller, & S. A. Jones (Eds.), Handbook of fat replacers (p. 336).
Boca Raton: CRC press.