**2. Materials and methods**

#### **2.1 Materials**

Bovine sodium caseinate powder, CMC low viscosity (50-200 cP, 4 % in H2O, 25 °C), GDL, tris(hydroxymethyl)aminomethane (Tris), 8-anilino-1-naphthalenesulfonate (ANS) as an ammonium salt, and sodium azide were purchased from Sigma-Aldrich Co. (Steinheim, Germany). HCl, and NaOH were provided by Cicarelli SRL (San Lorenzo, Argentina). The CMC stock solutions were also prepared in water and stored at 4 °C.

Acid-Induced Aggregation and Gelation of

vs. NaCAS concentration (wt%) curve.

**2.5 Acid aggregation** 

Bovine Sodium Caseinate-Carboxymethylcellulose Mixtures 79

The surface hydrophobicity (S0) was estimated according to the method of Kato and Nakai (Haskard & Li-Chan, 1998; Kato & Nakai, 1980), using ANS as an hydrophobic fluorescent marker. The measurements were carried out in an Aminco Bowman Series 2 spectrofluorometer, using an λex of 396 nm and an emission wavelength (λem) of 489 nm, previously determined from excitation and emission spectra of protein-ANS complex, at a constant temperature of 35ºC. The FI was measured in samples containing ANS 0.04 mM and consecutive aggregates of 0.1 wt% NaCAS with or without CMC (FIb). The FI was also determined in samples containing only protein (i.e. without the addition of the fluorescent probe) in the presence or absence of CMC at the same concentrations (FIp). The difference between FIb and FIp (F) was calculated and S0 was determined as the initial slope in the F

Kinetics of NaCAS aggregation induced by the acidification with GDL, in the presence or absence of CMC, was analyzed by measuring turbidity (τ) in the range of 450 to 650 nm, in a Spekol 1200 spectrophotometer with a diode arrangement and a thermostatized cell. The

wt% GDL <sup>R</sup>

R used for this experiment was 0.5, at a temperature of 35 °C. Acidification was initiated by

Changes in the protein average size were followed by the dependence of τ on wavelength

d log 4.2+ d log

 is a parameter that has a direct relationship with the average size of the particles, can be used to easily detect and follow rapid size changes, and was obtained from the slope of log τ versus log λ plots, in the 450 to 650 nm range, where the absorption due to the protein chromophores is negligible allowing the estimation of τ as absorbance (Camerini-Otero & Day, 1978; Risso et al., 2007). Absorption spectra and absorbance at 650 nm (A650) were registered as a function of time until a maximum and constant value of A650 was reached; simultaneously the pH decrease was measured. The measurements of pH were carried out on digital pH meter Orion SA 720, equipped with proton-selective glass membrane electrode combined with saturated calomel reference electrode. On the other hand, it has been shown that , for a system of aggregating particles of the characteristics of caseinates, tends, upon aggregation, toward an asymptotic value that can be considered as a fractal

To verify if β was actually related to the average size of the particles, the size distribution functions and the hydrodynamic diameters of NaCAS particles were determined by dynamic light scattering (DLS) using a Brookhaven 9863 Model equipment with a He–Ne laser (λ0 = 632.8 nm) with a maximal power of 15 mV, and using 90° as the measuring angle. Hydrodynamic diameters were calculated using the BI9000AT 6,5 Version software

  

wt% NaCAS (1)

(2)

amount of GDL added was calculated using the following relation:

the addition of solid GDL to 10 g of NaCAS suspension (0.5 wt%).

dimension (Df) of the aggregates (Horne, 1987; Risso et al., 2007).

**2.6 Changes in size and compaction of particles** 

(λ) of the suspensions, determined according to:

NaCAS suspensions were prepared from dissolution of commercial drug in distilled water (isoionic pH) at room temperature. After concentration measurements, 0.15 g.L-1 sodium azide was added as a bacteriostatic agent, and the solutions were stored at 4 °C. Protein concentration was determined by the Kuaye's method which is based on the ability of strong alkaline solutions to shift the spectrum of the amino acid tyrosine to higher wavelength values in the UV region (Kuaye, 1994).

Stock solutions nearly 6 mM for ANS were prepared in distilled water and stored in the dark at 4 °C; the concentration was determined by absorbance measurements, using a molar extinction coefficient (of 4,950 M-1 cm-1 at 350 nm.

#### **2.2 NaCAS–CMC phase diagram**

Phase diagram was established at pH 6.8 (buffer Tris-HCl 10 mM) and 35 °C, and it was constructed by determining the transition from single to two-phase systems. The thermodynamic compatibility study was carried out using the method proposed by Spyropoulos et al. They propose to carefully prepare a series of polysaccharide/protein aqueous solutions to give binary systems and incubate them a certain time at a given temperature, and then evaluate a single or two-phase systems formation (Spyropoulos et al., 2010).

These binary systems were prepared with the same CMC concentration but with NaCAS concentrations ranging from 0 to 4 wt% in one case, and with the same protein concentration but with polysaccharide concentrations ranging from 0 to 4.5 wt% in the other. A total of three samples were taken from each of these binary solutions and kept in sealed tubes at 35 °C for at least 24 h, after which the occurrence of phase separation (or not) was verified by visual inspection.

#### **2.3 Thermal stability of NACAS in the presence or absence of CMC**

The effects of thermal treatments on NaCAS, CMC and their mixtures were monitored through spectrophotometry with the aim of evaluating the biopolymer aggregation by heating at different temperatures. Measurements at increasing temperature were made at 650 nm from 10 to 100 °C with a heating rate of 0.5 °C per minute. The equipment used was a Jasco V-550 doublebean spectrophotometer equipped with a cell holder heated by Peltier effect and controlled by a programmable unit. The cell was filled with a 0.02 wt% NaCAS solution in buffer Tris-HCl 10mM pH 6.8 up to a final volume of 2.5 mL and sealed with a teflon stopper to avoid evaporation during each experiment. The spectrophotometer compartment was continuously purged with nitrogen to prevent the condensation of water vapor on the cell walls. This method was also performed in the presence of CMC at NaCAS:CMC proportion of 8:1, 4:1, 2:1, 1:1 and 1:1.5.

#### **2.4 Spectrofluorimetric determinations**

Fluorescence excitation and emission spectra of the NaCAS (0.1 wt%) were obtained using a spectrofluorometer Aminco Bowman Series 2. Measurements were carried out in the presence and absence of CMC, in order to detect any spectral shifts and/or changes in the relative intensity of fluorescence (FI). Previously, the excitation wavelength (λex) and the range of concentration with a non significant internal filter effect were determined. The samples (3 mL) used for the spectral analysis and FI measures were transferred into a fluorescence cell with a light path length of 1 cm and placed into a cell holder keeping the temperature constant at the fixed values desired. Values of FI were registered within the range of 300-400 nm at 35ºC using a λex of 286 nm.

The surface hydrophobicity (S0) was estimated according to the method of Kato and Nakai (Haskard & Li-Chan, 1998; Kato & Nakai, 1980), using ANS as an hydrophobic fluorescent marker. The measurements were carried out in an Aminco Bowman Series 2 spectrofluorometer, using an λex of 396 nm and an emission wavelength (λem) of 489 nm, previously determined from excitation and emission spectra of protein-ANS complex, at a constant temperature of 35ºC. The FI was measured in samples containing ANS 0.04 mM and consecutive aggregates of 0.1 wt% NaCAS with or without CMC (FIb). The FI was also determined in samples containing only protein (i.e. without the addition of the fluorescent probe) in the presence or absence of CMC at the same concentrations (FIp). The difference between FIb and FIp (F) was calculated and S0 was determined as the initial slope in the F vs. NaCAS concentration (wt%) curve.

#### **2.5 Acid aggregation**

78 Food Industrial Processes – Methods and Equipment

NaCAS suspensions were prepared from dissolution of commercial drug in distilled water (isoionic pH) at room temperature. After concentration measurements, 0.15 g.L-1 sodium azide was added as a bacteriostatic agent, and the solutions were stored at 4 °C. Protein concentration was determined by the Kuaye's method which is based on the ability of strong alkaline solutions to shift the spectrum of the amino acid tyrosine to higher

Stock solutions nearly 6 mM for ANS were prepared in distilled water and stored in the dark at 4 °C; the concentration was determined by absorbance measurements, using a molar

Phase diagram was established at pH 6.8 (buffer Tris-HCl 10 mM) and 35 °C, and it was constructed by determining the transition from single to two-phase systems. The thermodynamic compatibility study was carried out using the method proposed by Spyropoulos et al. They propose to carefully prepare a series of polysaccharide/protein aqueous solutions to give binary systems and incubate them a certain time at a given temperature, and then evaluate a single or two-phase systems formation (Spyropoulos et al.,

These binary systems were prepared with the same CMC concentration but with NaCAS concentrations ranging from 0 to 4 wt% in one case, and with the same protein concentration but with polysaccharide concentrations ranging from 0 to 4.5 wt% in the other. A total of three samples were taken from each of these binary solutions and kept in sealed tubes at 35 °C for at least 24 h, after which the occurrence of phase separation (or not) was verified by

The effects of thermal treatments on NaCAS, CMC and their mixtures were monitored through spectrophotometry with the aim of evaluating the biopolymer aggregation by heating at different temperatures. Measurements at increasing temperature were made at 650 nm from 10 to 100 °C with a heating rate of 0.5 °C per minute. The equipment used was a Jasco V-550 doublebean spectrophotometer equipped with a cell holder heated by Peltier effect and controlled by a programmable unit. The cell was filled with a 0.02 wt% NaCAS solution in buffer Tris-HCl 10mM pH 6.8 up to a final volume of 2.5 mL and sealed with a teflon stopper to avoid evaporation during each experiment. The spectrophotometer compartment was continuously purged with nitrogen to prevent the condensation of water vapor on the cell walls. This method was also performed in the presence of CMC at

Fluorescence excitation and emission spectra of the NaCAS (0.1 wt%) were obtained using a spectrofluorometer Aminco Bowman Series 2. Measurements were carried out in the presence and absence of CMC, in order to detect any spectral shifts and/or changes in the relative intensity of fluorescence (FI). Previously, the excitation wavelength (λex) and the range of concentration with a non significant internal filter effect were determined. The samples (3 mL) used for the spectral analysis and FI measures were transferred into a fluorescence cell with a light path length of 1 cm and placed into a cell holder keeping the temperature constant at the fixed values desired. Values of FI were registered within the

**2.3 Thermal stability of NACAS in the presence or absence of CMC** 

NaCAS:CMC proportion of 8:1, 4:1, 2:1, 1:1 and 1:1.5.

range of 300-400 nm at 35ºC using a λex of 286 nm.

**2.4 Spectrofluorimetric determinations** 

wavelength values in the UV region (Kuaye, 1994).

extinction coefficient (of 4,950 M-1 cm-1 at 350 nm.

**2.2 NaCAS–CMC phase diagram** 

2010).

visual inspection.

Kinetics of NaCAS aggregation induced by the acidification with GDL, in the presence or absence of CMC, was analyzed by measuring turbidity (τ) in the range of 450 to 650 nm, in a Spekol 1200 spectrophotometer with a diode arrangement and a thermostatized cell. The amount of GDL added was calculated using the following relation:

$$\text{IR} = \frac{\text{wt\%} \text{GDL}}{\text{wt\%} \text{NaCAS}} \tag{1}$$

R used for this experiment was 0.5, at a temperature of 35 °C. Acidification was initiated by the addition of solid GDL to 10 g of NaCAS suspension (0.5 wt%).

#### **2.6 Changes in size and compaction of particles**

Changes in the protein average size were followed by the dependence of τ on wavelength (λ) of the suspensions, determined according to:

$$\beta = 4.2 + \frac{\mathrm{d}(\log \tau)}{\mathrm{d}(\log \lambda)}\tag{2}$$

 is a parameter that has a direct relationship with the average size of the particles, can be used to easily detect and follow rapid size changes, and was obtained from the slope of log τ versus log λ plots, in the 450 to 650 nm range, where the absorption due to the protein chromophores is negligible allowing the estimation of τ as absorbance (Camerini-Otero & Day, 1978; Risso et al., 2007). Absorption spectra and absorbance at 650 nm (A650) were registered as a function of time until a maximum and constant value of A650 was reached; simultaneously the pH decrease was measured. The measurements of pH were carried out on digital pH meter Orion SA 720, equipped with proton-selective glass membrane electrode combined with saturated calomel reference electrode. On the other hand, it has been shown that , for a system of aggregating particles of the characteristics of caseinates, tends, upon aggregation, toward an asymptotic value that can be considered as a fractal dimension (Df) of the aggregates (Horne, 1987; Risso et al., 2007).

To verify if β was actually related to the average size of the particles, the size distribution functions and the hydrodynamic diameters of NaCAS particles were determined by dynamic light scattering (DLS) using a Brookhaven 9863 Model equipment with a He–Ne laser (λ0 = 632.8 nm) with a maximal power of 15 mV, and using 90° as the measuring angle. Hydrodynamic diameters were calculated using the BI9000AT 6,5 Version software

Acid-Induced Aggregation and Gelation of

**2.8 Statistical analysis** 

**3. Results and discussions** 

phase samples, (●) two-phase gel-like systems

absence of the polysaccharide.

**3.2 Thermal stability of NACAS: CMC mixtures** 

system resolution for the protein gels images was:

Bovine Sodium Caseinate-Carboxymethylcellulose Mixtures 81

the pixel width in m, linear calibration was carried out using a micrometer rule. The final

The data are reported as the average values ± their standard deviations. Statistical analysis was performed with Sigma Plot 10.0 and Image J softwares. Relationship between variables was statistically analyzed by correlation analysis using Pearson correlation coefficient (r).

The results obtained for mixtures of NACAS and CMC are shown in Fig. 1. The polysaccharide and protein concentrations, in each of the prepared binary solutions, correspond to a single point on the phase diagram. This approach provides a "map'' of the

Fig. 1. Approach used for the determination of the phase diagrams for NaCAS:CMC systems after 24 h at 35 °C. Key: (○) one-phase clear solution, (∆) one-phase turbid solution, (▲) two-

Both the CMC as all mixtures NaCAS:CMC in all relations tested were not affected by rising temperature within the temperature range studied (10-100 °C). This would indicate that the polysaccharide is thermally stable in this range, and that the addition of CMC to NACAS increases its thermal stability, since the NaCAS starts aggregating at about 60 °C in the

*m Resolution* 15.5 *pixel m*

(4)

1pixel width 0.0645 0.0005

The differences were considered statistically significant at p < 0.05 values.

transition from the single-phase to the two-phase region of the phase diagram.

**3.1 Thermodynamic compatibility of NaCAS:CMC mixtures** 

processing. To carry out this determination an amount of solid GDL was added to 8 mL of a 0.5 wt% NaCAS solution in order to obtain a GDL/protein relation of 0.5. Measurements at different times were performed until the maximum of allowed by the instrument was reached while pH was simultaneously monitored.

#### **2.6.1 Effect of CMC on the viscosity of media**

The aggregation process is limited by diffusion, which depends on the medium viscosity (). Therefore, is important to determine the effect that the presence of the polysaccharide exerts on that property. The was measured in triplicate, using a rotational viscosimeter Brookfield LV Master (LVDV-III) with cone/plate geometry and thermostatically controlled at a temperature of 35.00 ± 0.05ºC. The relative viscosity (r) was calculated as:

$$
\eta\_{\mathbf{r}} = \eta\_{\text{sol}} / \eta\_0 \tag{3}
$$

where sol is the solution viscosity and 0 is the water ones.

#### **2.7 Acid gelation**

Above a certain protein concentration, the loss of electrostatic stability by acidification results in the formation of a three-dimensional gel network. Effect of CMC concentration on the kinetic of gelation, rheological properties and microstructure of gels were investigated.

#### **2.7.1 Rheological properties of acid gels**

Rheological properties of NaCAS samples (3 wt%), in the absence or presence of CMC, were determined in a stress and strain controlled rheometer AR G2 model using a cone geometry (diameter: 40 mm, cone angle: 2°, cone truncation: 55 mm) and a system of temperature control with a recirculating bath (Julabo model ACW 100) connected to a Peltier plate. An amount of solid GDL according to a certain R was added to initiate the acid gelation.

Measurements were performed each 20 sec with a constant oscillation stress of 0.1 Pa and a frequency of 0.1 Hz. The Lissajous figures at various times were plotted to ensure that the measurements of storage or elastic modulus (G') and loss or viscous modulus (G'') were always obtained within the linear viscoelastic region.

The G'-G'' crossover times (tg) of acidified caseinate systems were considered here as the gel times, since most studies of milk/caseinate gelation have adopted this criterion (Braga et al., 2006; Curcio et al., 2001). pH at tg was also determined considering the pH value at the G'- G'' crossover (pHg).

#### **2.7.2 Conventional inverted microscopy**

The degree of compactness of gels was evaluated through digital image analysis. For this, bottom surface image of gel were obtained by conventional inverted microscopy. To obtain the microscopic images, 90 μL of each sample were placed in compartments of the LAB-TEK II cells. The samples were obtained by duplicate under a constant temperature set at 35ºC. Transmission images of gels were obtained using a conventional inverted microscopy (Union Optical) with an objective 100x and a digital camera (Canon PowershotA640) with a zoom 7.1x and microscope adapter of 52 mm.

The average pore diameters of gels were determined using the program Image J. To do this, straight lines were drawn on the digital images and values of pore size were measured in pixels. These values were averaged (n=5) and obtained the average pore size. To determine the pixel width in m, linear calibration was carried out using a micrometer rule. The final system resolution for the protein gels images was:

$$1\,\text{pixel width} = \left(0.0645 \pm 0.0005\right)\mu m \Rightarrow \text{Resolution} = 15.5\,\text{pixel}/\mu m \tag{4}$$

### **2.8 Statistical analysis**

80 Food Industrial Processes – Methods and Equipment

processing. To carry out this determination an amount of solid GDL was added to 8 mL of a 0.5 wt% NaCAS solution in order to obtain a GDL/protein relation of 0.5. Measurements at different times were performed until the maximum of allowed by the instrument was

The aggregation process is limited by diffusion, which depends on the medium viscosity (). Therefore, is important to determine the effect that the presence of the polysaccharide exerts on that property. The was measured in triplicate, using a rotational viscosimeter Brookfield LV Master (LVDV-III) with cone/plate geometry and thermostatically controlled

> r 0

Above a certain protein concentration, the loss of electrostatic stability by acidification results in the formation of a three-dimensional gel network. Effect of CMC concentration on the kinetic of gelation, rheological properties and microstructure of gels were investigated.

Rheological properties of NaCAS samples (3 wt%), in the absence or presence of CMC, were determined in a stress and strain controlled rheometer AR G2 model using a cone geometry (diameter: 40 mm, cone angle: 2°, cone truncation: 55 mm) and a system of temperature control with a recirculating bath (Julabo model ACW 100) connected to a Peltier plate. An

Measurements were performed each 20 sec with a constant oscillation stress of 0.1 Pa and a frequency of 0.1 Hz. The Lissajous figures at various times were plotted to ensure that the measurements of storage or elastic modulus (G') and loss or viscous modulus (G'') were

The G'-G'' crossover times (tg) of acidified caseinate systems were considered here as the gel times, since most studies of milk/caseinate gelation have adopted this criterion (Braga et al., 2006; Curcio et al., 2001). pH at tg was also determined considering the pH value at the G'-

The degree of compactness of gels was evaluated through digital image analysis. For this, bottom surface image of gel were obtained by conventional inverted microscopy. To obtain the microscopic images, 90 μL of each sample were placed in compartments of the LAB-TEK II cells. The samples were obtained by duplicate under a constant temperature set at 35ºC. Transmission images of gels were obtained using a conventional inverted microscopy (Union Optical) with an objective 100x and a digital camera (Canon PowershotA640) with a

The average pore diameters of gels were determined using the program Image J. To do this, straight lines were drawn on the digital images and values of pore size were measured in pixels. These values were averaged (n=5) and obtained the average pore size. To determine

amount of solid GDL according to a certain R was added to initiate the acid gelation.

*sol* (3)

at a temperature of 35.00 ± 0.05ºC. The relative viscosity (r) was calculated as:

reached while pH was simultaneously monitored.

**2.6.1 Effect of CMC on the viscosity of media** 

**2.7.1 Rheological properties of acid gels** 

always obtained within the linear viscoelastic region.

**2.7.2 Conventional inverted microscopy** 

zoom 7.1x and microscope adapter of 52 mm.

**2.7 Acid gelation** 

G'' crossover (pHg).

where sol is the solution viscosity and 0 is the water ones.

The data are reported as the average values ± their standard deviations. Statistical analysis was performed with Sigma Plot 10.0 and Image J softwares. Relationship between variables was statistically analyzed by correlation analysis using Pearson correlation coefficient (r). The differences were considered statistically significant at p < 0.05 values.
