Original papers
Physical Properties of Heat-induced Whey Protein Aggregates Formed at pH 5.5 and 7.0
2017 Volume 23 Issue 4 Pages 595-601
Details
2017 Volume 23 Issue 4 Pages 595-601
Effects of pH on the physical properties of heat-induced whey protein aggregates were investigated using atomic force microscopy. At pH 5.5 and a protein concentration of 0.3% (w/w), micrometer-sized aggregates were formed within 1 min of heating at 80°C, while it took more than 30 min for aggregates to reach similar sizes at pH 7.0 and a protein concentration of 2.0% (w/w). The surface roughness of air-dried aggregates evaluated based on cross-sectional height data decreased by a factor of 2 with increasing pH from 5.5 to 7.0. Young's modulus values of whey protein aggregates evaluated based on force-indentation curve measurements without drying decreased by a factor of 3 with increasing pH from 5.5 to 7.0. The present results suggest that heat-induced whey protein aggregates became mechanically weaker with increasing pH due to increased inter-/intra-molecular electrostatic repulsions.
Whey proteins provide not only numerous physiological benefits, but also physicochemical functions such as foaming, emulsifying, and gelling abilities (Foegeding, 2015; Lagrange et al., 2015). Whey proteins are classified as globular proteins whose polypeptide chains are folded into specific conformations in the native state (de la Fuente et al., 2002; Nicolai and Durand, 2007; Mehalebi et al., 2008a). Upon heating at above denaturation temperatures, whey proteins unfold at least partially to expose hydrophobic cores and thiol groups that are reactive around neutral pH, and form aggregates through hydrophobic interactions and covalent disulfide bond formation (Jung et al., 2008; Mezzenga and Fischer, 2013). Whey proteins may function as a texture modifier through the formation of such heat-induced aggregates, since the effective hydrodynamic volume of an aggregate can be significantly larger than the sum of those of individual molecules forming the aggregate (Foegeding et al., 2002; Purwanti et al., 2011).
It is known that pH is a critical factor in determining the structure and functionality of heat-induced whey protein aggregates (Langton and Hermansson, 1992; Hudson et al., 2000). As pH decreases from neutral pH, thiol groups become less reactive and hydrophobic interactions play more predominant roles in the formation of heat-induced aggregates (Bromley et al., 2006; Donato et al., 2009; Mehalebi et al., 2008b; Zuniga et al., 2010). Particulate aggregates are formed at pH between 4 and 6, which is around the isoelectric point of whey protein (i.e., pH 5.2), while fine structures are formed at pH less than 4 or greater than 6 (Langton and Hermansson, 1992; Aymard et al., 1999; Ikeda and Morris, 2002; Jung et al., 2008). In gel fracture tests, gels composed of particulate aggregates or fine aggregates formed at pH greater than 6 showed relatively large stresses and strains at fracture (Hudson et al., 2000). In contrast, gels composed of fine aggregates formed at pH less than 4 exhibited fairly low fracture stresses and strains, indicating the mechanical weakness of these aggregates (Hudson et al., 2000).
Microparticulation is a widely adopted technology for modifying whey protein functionality in order to enhance their utilization as a fat replacer, which imparts fat-like textural characteristics to reduced fat foods (Torres et al., 2011; Chung et al., 2014; Sturaro et al., 2015). Microparticulated whey protein particles have a spherical shape and a size range similar to those of lipid droplets in food emulsions (Singer, 1996). Such protein particles are thought to roll freely over one another under shear flow and provide a smooth and creamy mouthfeel (Liu et al., 2015; 2016a; 2016b). The conventional method of microparticulation involves the heat-induced aggregation of whey proteins in combination with high shear to produce micrometer-sized spherical particles of whey proteins (Singer, 1996), while novel methods for achieving greater control over the structure and functionality of microparticulated whey protein particles are emerging. Micrometer-sized spherical particles of whey proteins were prepared by heating water-in-oil emulsions consisting of aqueous whey protein solution droplets dispersed in continuous oil phases (Saglam et al., 2011; 2012; 2014). This approach was modified to incorporate whey protein solutions in microemulsions and produce nanometer-sized whey protein particles (Zhang and Zhong, 2010). Spherical whey protein nanoparticles were also obtained by heating dilute aqueous solutions of whey protein in a narrow pH range (i.e., 5.8 – 6.2) around the isoelectric point of the protein (Schmitt et al., 2009, 2010; Phan-Xuan et al., 2011).
Tribological and rheological studies of microparticulated whey protein particles dispersed in liquids and gels showed that the addition of whey protein particles to liquids decreased friction coefficients mainly due to ball-bearing lubrication and that the reduction of friction of gels containing whey protein particles was more complex and dependent on various factors including bulk viscosity, gel fracture properties, and interactions between whey protein particles and gel matrices (Liu et al., 2016b). In order to better understand the structure and functionality relationship of microparticulated whey protein particles, the physical properties of their dispersions as well as those of individual particles require elucidation. The objective of this study was to obtain micrometer-sized whey protein particles by heating dilute aqueous solutions of whey protein adjusted to pH 5.5 – 7.0 and to characterize their physical properties using atomic force microscopy (AFM).
Materials Whey protein isolate (WPI) (BiPRO, lot JE 226-2-420) was supplied by Davisco Foods International Inc. (Le Sueur, MN, USA). The protein content of the dry powder was >90% (w/w) according to the manufacturer's specifications. All analytical grade reagents were purchased from Fisher Scientific Co. (Pittsburgh, PA, USA). Mica sheets were purchased from Electron Microscopy Sciences (Hartfield, PA, USA).
Preparation of heat-induced whey protein aggregates Whey protein powders were dispersed in distilled water containing 0.1 M NaCl, stirred overnight using a magnetic stirrer, and adjusted to pH 3 – 7 by adding small amounts of HCl and to protein concentrations of 0.3, 1.2, or 2.0% (w/w) by adding small amounts of 0.1 M NaCl aqueous solution. The 0.1 M NaCl solution was used to prepare whey protein solutions in order to make differences in the ionic strength resulting from the addition of various amounts of HCl for pH adjustment negligible. Protein solutions were then transferred to sealed glass vials and heated in a water bath maintained at 80 ± 2°C. Subsamples were taken at pre-specified time intervals and quenched in an ice water bath.
Particle size analysis Particle sizes of heat-induced whey protein aggregates were measured at room temperature using a Zetasizer Nano-ZS (Malvern Instruments Ltd., Malvern, UK). The instrument measured the autocorrelation function of the intensity of light scattered from particles moving in Brownian motion to calculate the diffusion coefficient subsequently used to calculate particle diameter based on the Stokes-Einstein relationship. A relative refractive index value of the particle of 1.52 and that of the solvent of 1.33 were used for the calculation.
AFM Sample solutions containing heat-induced whey protein aggregates were diluted to give a protein concentration of 10 – 100 ppm. Aliquots (2 µL) of diluted samples were deposited onto freshly cleaved mica surfaces, air-dried for 15 min, and imaged in air using a BioScope Catalyst atomic force microscope (Bruker Corporation, Santa Barbara, CA, USA) operated in peak force tapping mode. Mica surfaces were scanned at a rate of 1 Hz using a V-shaped silicon nitride cantilever probe with a nominal spring constant of 0.4 N;·m−1 and a resonant frequency of approximately 70 kHz (Bruker Corporation, Camarillo, CA, USA). Topographical images were generated based on the trace of the z-position of the cantilever at its maximum deflection, complemented by the equivalent peak force error image generated by recording slight fluctuations of the peak force around the preset value during scanning, and analyzed using NanoScope Analysis software version 1.40 (Bruker Corporation, Santa Barbara, CA, USA). In order to evaluate Feret's diameter, representing the diameter of the smallest circle that fully covers a protein aggregate, topographical images were converted to binary images and analyzed using ImageJ software version 1.49 (National Institutes of Health, Bethesda, MD, USA).
The Young's modulus of heat-induced whey protein aggregates was evaluated using the microscope operated in quantitative nanomechanical mapping mode. Whey protein aggregates were immobilized onto poly-L-lysine coated mica surfaces without drying, placed in a fluid cell filled with a buffer solution, and scanned using a V-shaped silicon nitride cantilever probe having a nominal spring constant of 0.7 N·m−1 and a resonant frequency of approximately 150 kHz (Bruker Corporation, Camarillo, CA, USA). At least 30 force (F)-indentation (δ) curves were obtained for each whey protein aggregate and analyzed to calculate the Young's modulus (E) based on the Hertz model frequently used to study biological materials (Uricanu et al., 2004).
 |
where v is the Poisson's ratio and r is the tip radius.
Statistical analysis Experiments were performed in triplicate to report means and standard deviations. Analysis of variance (ANOVA) and Tukey's HSD test were performed using SigmaPlot software version 13.0 (Systat Software Inc., San Jose, CA, USA). P-values less than 0.05 were considered to be statistically significant.
Size developments during heat-induced aggregation Figure 1 shows changes in the size distribution of whey protein aggregates during heating at pH 5.5 and a protein concentration of 1.2% (w/w). After heating for 4 min at 80°C, a bimodal distribution with a peak around 60 – 70 nm and another peak around 300 nm was observed. These peaks shifted to larger diameters with increasing heating period to 8 min. After heating for 12 min, the first peak in the smallest diameter range shifted to ca. 250 nm, the second peak in a larger diameter range decreased its area without shifting, and a third peak truncated at 5 µm due to the detection limit of the instrument appeared in the micrometer range. The first and second peaks did not shift after longer periods of heating up to 20 min, but decreased their areas with increasing heating periods. The third peaks in the micrometer range appeared to increase their areas with increasing heating periods. These results suggest that the heat-induced aggregation of whey protein is unlikely to be a simple nucleation and growth process resulting in a monomodal size distribution (Bromley et al., 2006). Aymard et al. (1996) observed two populations of whey protein aggregates prepared by heating at neutral pH in size-exclusion chromatography after elution of unaggregated proteins, and suggested that the aggregation occurred in two steps, where small globular aggregates having a diameter of approximately 30 nm were formed first and then they aggregated to form larger clusters. Later light scattering studies showed that the two-step aggregation model was applicable to the heat-induced aggregation of whey protein at a lower pH range between 5.2 and 7 (Mehalebi et al., 2008b). Consistent with these previous studies, the present results indicate that the heat-induced aggregation of whey protein at pH 5.5 and 7.0 involves multiple steps, resulting in the formation of intermediate aggregates.
Changes in particle size distributions in 1.2% (w/w) whey protein solutions during heating at 80°C and pH 5.5 for 4 min (●), 8 min (○), 12 min (▴), 16 min (△), and 20 min (■).
Figure 2 shows developments of the mean diameter of whey protein aggregates during heating at pH 5.5 – 7.0 and protein concentrations of 0.3 – 2.0% (w/w). At a protein concentration of 0.3% (w/w), the mean diameter appeared to reach a plateau or did not change significantly with time in all examined pH conditions, indicating that the size of aggregates was limited by the low protein concentration. At a protein concentration of 1.2% (w/w), the mean diameter exceeded 300 nm within 4 min of heating at pH 5.5, while it took more than 40 min at pH 6.3 and 100 min at pH 7.0 for the mean diameter to reach 200 nm. At the highest protein concentration of 2% (w/w), the mean diameter exceeded 600 nm within 4 min of heating at pH 5.5, while it took 20 min at pH 6.3 and 75 min at pH 7 for aggregates to develop to a similar size. At all examined protein concentrations, the heat-induced aggregation of whey protein was fastest at pH 5.5. Zuniga et al. (2010) reported that values of the apparent reaction rate constant of heat-induced aggregation of β-lactoglobulin increased from 5.6×10−3 to 7.5×10−3 (s−1) with decreasing pH from 6.8 to 6. An increase in the reaction rate of heat-induced aggregation of β-lactoglobulin with decreasing pH was also reported in the pH range between 5.7 and 5.9 (Donato et al., 2009). Vreeker et al. (1992) conducted static light scattering studies of heat-induced whey protein aggregation in an intermediate pH range between 4 and 6. The results suggested that aggregation at pH 4.9 was diffusion-limited but that at pH 5.4 was reaction-limited, where the probability of sticking of two protein particles upon contact is relatively low due to the presence of electrostatic repulsive forces acting between particles. It is therefore likely that the fastest aggregation rate of heat-denatured whey protein at pH 5.5 observed in this study is due to the net charge of the protein being lowest at the pH near its isoelectric point, resulting in fewer electrostatic repulsions between protein molecules compared with those at higher pH conditions.
Developments of mean diameters in 0.3% (w/w) (a), 1.2% (w/w) (b), and 2.0% (w/w) (c) whey protein solutions during heating at 80°C and pH 5.5 (●), 6.3 (○), or 7.0 (▴).
Physical properties of heat-induced whey protein aggregates The heat-induced aggregation of the major whey protein β-lactoglobulin at neutral pH has been proposed as a two-step process based on scattering studies (Aymard et al., 1996). The average hydrodynamic radius of aggregates increased with time until reaching a plateau value of ca. 16 nm at a low ionic strength, while these primary aggregates further aggregated into large fractal clusters at higher ionic strengths (Aymard et al., 1996). Later AFM studies revealed that the size of primary aggregates formed by heating aqueous solutions of WPI increased from 11 to 27 nm with increasing concentrations of added NaCl from 0 to 0.3 mol·L−1 (Ikeda and Morris, 2002). Figure 3 shows representative peak force error images of heat-induced whey protein aggregates formed at pH 5.5 and 7.0, deposited on mica surfaces, and air-dried before imaging. The formation of micrometer-sized aggregates was detected in AFM images of samples heated at 80°C for less than 1 min at pH 5.5 and more than 30 min at pH 7.0. These aggregates appeared to consist of elementary particles larger than individual protein molecules regardless of pH conditions, suggesting that the two-step aggregation process reported originally for the heat-induced aggregation of whey proteins at neutral pH also took place at pH 5.5. This observation is consistent with the particle size distribution in heat-induced whey protein aggregates formed at pH 5.5 (Fig. 1) in that the formation of intermediate aggregates is indicated.
Peak force error images of heat-induced whey protein aggregates formed at pH 5.5 and a whey protein concentration of 0.3% (w/w) (a–d) or at pH 7.0 and a whey protein concentration of 2.0% (w/w) (e–h). Whey protein solutions were heated at 80°C for 1 min (a), 3 min (b), 7 min (c), 10 min (d), 20 min (e), 30 min (f), 45 min (g), and 60 min (h). Scale bars represent 1 µm.
The surfaces of aggregates formed at pH 7.0 appear to be smoother than those formed at pH 5.5 in Fig. 3. In order to quantify the roughness of aggregate surfaces, the number of peaks along the aggregate surface was evaluated. Thirty aggregates whose diameters ranged from ca. 1 – 2 µm were randomly selected at each pH to ensure unbiased measurement and their cross-sectional height profiles were analyzed. Figure 4 shows that the number of peaks decreases as pH increases from pH 5.5 to 7.0 by a factor of 2, indicating that aggregates prepared at pH 7.0 have been flattened to a greater degree on the mica upon air-drying than those prepared at pH 5.5. Saglam et al. (2012) reported similar results showing that heat-induced particles of WPI prepared at pH 6.8 collapsed after being air-dried for scanning electron microscopy studies but those prepared at pH 5.5 remained intact, suggesting that particles formed at pH 5.5 were mechanically stronger than those formed at pH 6.8 and could resist the interfacial tension imposed at the air-liquid phase boundary during air-drying. At pH 6.8, whey protein molecules are net-negatively charged and thiol groups are reactive, while the heat-induced aggregation of whey protein is mainly driven through hydrophobic interactions at pH 5.5 (Jung et al., 2008; Mezzenga and Fischer, 2013). At the latter pH, electrostatic attractive forces between oppositely charged groups may also play certain roles even though the net charge is close to zero (Majhi et al., 2006).
Numbers of peaks per cross-section of heat-induced whey protein aggregates formed at pH 5.5 – 7.0. Different letters indicate statistically significant differences (P < 0.05).
Rheological properties of dispersions and suspensions of particles are influenced by many factors including the particle shape (Wolf et al., 2001; Akkermans et al., 2008). In order to evaluate the anisotropy or elongation of heat-induced whey protein aggregates formed in different pH conditions, the area and Feret's diameter of each aggregate were evaluated using AFM images since the area-to-Feret's diameter ratio is more sensitive to elongation than other types of shape parameters such as the perimeter ratio (d'Almeida and Paciornik, 2007). In Fig. 5, areas occupied by individual aggregates in AFM images are plotted as a function of Feret's diameter. If aggregates in AFM images are circular, then the area should be related to the square of Feret's diameter. As the particle shape becomes more anisotropic, the area would be less dependent on Feret's diameter. The slopes of the plots shown in Fig. 5 increased from 1.4 at pH 3, to 1.6 at pH 4, and to 1.9 at pH 5.5 – 7.0, indicating that heat-induced whey protein aggregates were more extended and anisotropic in shape at pH 3 and 4, and that their overall shapes were fairly spherical at pH 5.5 – 7.0 regardless of pH conditions. It was therefore considered that the mechanical properties of heat-induced whey protein particles would play more significant roles in the determination of their texture modifying abilities.
Relationships between the area and Feret's diameter of heat-induced whey protein aggregates formed at pH 3 (● in a), 4 (○ in a), 5.5 (● in b), 6.3 (○ in b), and 7.0 (△ in b).
Young's modulus values of heat-induced aggregates of whey proteins derived from force-indentation curves are shown in Fig. 6. Heat-induced aggregates prepared at pH 5.5 and 6.3 at three different protein concentrations showed similar Young's modulus values of around 25 – 30 MPa, while those prepared at pH 7.0 showed significantly smaller modulus values of less than 15 MPa. Young's modulus values of aggregates prepared at pH 7 and protein concentrations of 1.2% (w/w) and 2.0% (w/w) were also not significantly different, while they were approximately three times that of aggregates prepared at the same pH and a protein concentration of 0.3% (w/w). These results are consistent with the results from the evaluation of the roughness of air-dried aggregate surfaces (Fig. 4), where aggregates prepared at pH 7.0 are suggested to be mechanically weaker than those prepared in other pH conditions. Uricanu et al. (2004) evaluated Young's modulus values of casein micelles and those of their aggregates based on force-indentation curve measurements, and found that the former values were approximately 3 orders of magnitude greater, indicating that the mechanical strength of casein micelle aggregates was determined largely by that of inter-micellar links rather than that of casein micelle itself. The present results suggest that whey protein molecules in heat-induced aggregates formed at pH 7.0 are less tightly associated with each other than those formed in lower pH conditions, presumably due to more enhanced intermolecular electrostatic repulsions, even though the formation of intermolecular disulfide bonds is expected to be further enhanced at pH 7.0. More enhanced intermolecular electrostatic repulsions are also expected to reduce the degree of intermolecular hydrophobic interactions.
Young's moduli of heat-induced whey protein aggregates formed at pH 5.5 – 7.0 and protein concentrations of 0.3% (w/w) (●), 1.2% (w/w) (○), and 2.0% (w/w) (▴). Different letters indicate statistically significant differences (P < 0.05).
Atomic force microscopy was demonstrated to be a useful tool for the study of not only the structure but also the physical properties of micrometer-sized whey protein aggregates formed by heating in various pH conditions. The rate of heat-induced aggregation increased significantly with decreasing pH from 7.0 to 5.5, while the aggregation process appeared to involve the formation of primary aggregates and secondary aggregation between those primary aggregates regardless of pH conditions. The overall shape of micrometer-sized whey protein aggregates appeared to be minimally dependent on pH in the pH range between 5.5 and 7.0. Heat-induced whey protein aggregates formed at pH 7.0 collapsed upon air-drying, indicating mechanical weaknesses of these aggregates. This hypothesis was supported by the results from Young's modulus determinations of aggregates based on force-indentation curve measurements. The present results suggest that heat-induced whey protein aggregates became more swollen and mechanically weaker with increasing pH from 5.5 to 7.0 due to increased electrostatic repulsions between protein molecules. Consequently, the manipulation of pH during heat-induced aggregation of whey protein is suggested to be a promising approach to controlling the texture modifying abilities of whey protein microparticles.
Acknowledgements We thank Davisco Foods International, Inc. (Le Sueur, MN, USA) for kindly providing the whey protein isolate samples. This work was partially supported by the USDA National Institute of Food and Agriculture (Washington DC, USA; Hatch project 231867).
