Acid gelation of whey protein microbeads of different sizes
In acidified dairy products, the size of the whey protein particles could play a key role in the final structure of the gel. In the present study, small (SM; 2.5 ± 1.2 μm), medium (MM; 4.2 ± 2.2 μm), and large (LM; 18.4 ± 7.2 μm) whey protein microbeads were produced by mixing a 150 g.kg−1 whey protein isolate (WPI) solution and n-dodecane in the presence of polyglycerolpolyricinoleate (PGPR) surfactant at different shear rates and were then stabilized through heat gelation. The microbeads were then washed by centrifugation, dispersed at 70 or 90 g.kg−1 in milk ultrafiltrate, and acidification was performed at 35 °C by adding glucono-δ-lactone to achieve the final pH of ~4.5 in 6 h. Acid gelation was monitored using small deformation rheology, while the gel microstructure was investigated microscopically. The results showed that smaller size of microbeads promoted gels with a higher stiffness and a smaller pore size distribution. The effects were particularly significant at SM microbeads as the number of particles in this system was higher than in LM or MM, hence more connectivity between particles.
KeywordsWhey proteins Microbeads Size Acid gel Rheology Microstructure
The design of nanoparticles or microparticles of whey proteins with tailored geometries and physicochemical properties has received great interest in the latest decade (Nicolai et al. 2011; Sağlam et al. 2014c).Numerous applications are envisaged, especially for the control of rheological properties in food (Purwanti et al. 2012; Giroux et al. 2015; Westerik et al. 2015) including fat replacement in dairy foodstuff (Janhoj and Ipsen 2006; Celigueta Torres et al. 2009) or the improvement of emulsion and/or particle stability (Sağlam et al. 2014a; Çakır-Fuller 2015; Zhang and Zhong 2010). Recent prospects also investigate the encapsulation of bioactive or sensible compounds, or of living micro-organisms, into edible particles (Giroux and Britten 2011; Doherty et al. 2012; Sadeghi et al. 2014).
Essentially, three technological approaches are used to produce such microparticles (Sağlam et al. 2014a). Earlier strategies used the heat denaturation of the whey protein solutions in various conditions of protein concentration, pH, and ionic environment, also combined with shear or homogenization during or after heat-treatment to breakdown particles. However, modification of concentration, pH, or ionic environment also affects the aggregation mechanism, the density and/or the internal structure of the aggregates, the involvement of ions or the thiol/disulfide balance, which can all also affect the gelation properties (Morand et al. 2011a; Nicolai et al. 2011). A more subtle approach is to use thermodynamic equilibrium to coacervate whey proteins into a separated phase of high-protein microbeads (Bouhallab and Croguennec 2013), with possible sophistication involving addition of inert biopolymers and/or heat (Sağlam et al. 2014a). As resulting from equilibrium, the particles are often not stable upon e.g. acidification but raise great interest a means to produce switchable vehicles to deliver bioactives upon variation of the environmental conditions. The third approach, microemulsification, is a drastic—and historical—form of the latter, where an aqueous protein solution is dispersed and stabilized into oil in the presence of surfactant molecules. A major interest of microemulsions is the possibility to control size and roughness of spherical particles with possible decoupling of internal composition and structure (Zhang and Zhong 2010; Saglam et al. 2012; Westerik et al. 2015).
In the present research, we aimed at investigating the role of the size of heat-denatured whey protein particles on acid gelation, in the general context of understanding yoghurt-making mechanisms. To do this, heat-denatured whey protein particles were needed, with varying sizes but with essentially constant other physicochemical properties, both on the surface and inside.
To date, little research has been published on the use of controlled size whey protein microparticles in order to control acid gel texture. Ju and Kilara (1998) or Britten and Giroux (2001) used variation in calcium concentration and/or pH to produce whey protein aggregates of various sizes from heated dilute solutions and found that largest aggregates yielded gels with lowest firmness, hardness, and/or elasticity. Celigueta Torres et al. (2009) showed that it was possible to produce microparticulated whey protein particles with different sizes through heat-shear process and showed that acid-gel viscosity was inversely correlated with size. However, in these studies, many other factors were at play, e.g., protein concentration (Ju and Kilara 1998), shape and internal structure (Britten and Giroux 2001), and/or the amount of native proteins (Torres et al. 2011).
Hence, the first objective of this study was to describe the formation of whey protein particles of different sizes but constant other physicochemical properties, having a defined spherical geometry, called whey protein “microbeads.” The second objective was then to evaluate the effect of their size on the acid gelation. To do this, emulsification was used to create a water in oil (w/o) emulsion, in which the WPI solution was used as the dispersed phase and using a high speed mixer in the presence of an oil-soluble emulsifier (Saglam et al. 2012). Therefore, in the present study, whey protein microbeads of different sizes were produced, stabilized through heat gelation, then washed, and redispersed in milk ultrafiltration permeate to assess their acid gelation at a constant total protein concentration (i.e., constant volume fraction).
2 Materials and methods
A commercial whey protein isolate (WPI) (Prolacta 95, Lactalis Ingredients, Bourgbarré, France) was used as the source of protein. The protein content of WPI powder was ~939 g.kg−1 dry matter determined by the Kjeldahl method using 6.38 as conversion factor. Milk ultrafiltration permeate (MUF) containing the aqueous phase of milk (water and dialyzable solutes such as lactose and ions) was produced from microfiltration then ultrafiltration of raw skim milk (Schuck et al. 1994) and stored at 5 °C after addition of 0.2 g.L−1 sodium azide (NaN3) to ensure biological stability during experiments. Polyglycerolpolyricinoleate (Grindsted PGPR 90, Denmark) was purchased from Danisco and contained a polyglycerol ester of polycondensedricinoleic acid with added antioxidants α-tocopherol (E 307; max 200 ppm) and citric acid (E330; max 200 ppm). n-Dodecane (non-food-grade) was purchased from Merck, having a density of 0.748 kg.L−1 at 20 °C. All other chemicals were of analytical grade quality. PGPR is a food-grade surfactant that has proven efficient of the proposed approach (Saglam et al. 2012). The present research pursue academical objectives but may open perspectives for possible application providing that n-dodecane is replaced and/or that nonprotein additives are entirely removed, as in Zhang and Zhong (2010).
2.2 Formation of whey protein microbeads by emulsification
WPI solution at 150 g.kg−1 of protein was produced by diluting the WPI powder in deionized water with 0.2 g.kg−1 NaN3, stirring for 2 h, filtering on 0.45-μm hydrophilic syringe filter (Minisart, Sartorius syringe filter) and then left at 4 °C overnight for complete hydration. PGPR (25 g.kg−1) was dissolved in n-dodecane by stirring for 1 h at 40 °C. A water in n-dodecane (w/o) emulsion (20 g) was prepared by mixing the WPI solution (2 g) with n-dodecane plus PGPR (18 g), using a high speed mixer (Ultraturrax T18 basic, IKA Werke, Germany). According to Wilson et al. (1998), Kobayashi et al. (2005), and Saglam et al. (2012), the presence of PGPR is important to help forming and stabilizing the whey protein microdroplets in the hydrophobic continuous phase by lowering the interfacial tension, and hence the stress needed to break up droplets into smaller particles is reduced. Surfactants also prevent coalescence of newly formed microbeads (Tadros 2013). Different sizes of whey protein microdroplets were produced by mixing the w/o emulsions at different speeds, i.e., 7,200, 15,600, and 24,200 rpm for 5 min, and were called large (LM), medium (MM), and small (SM) size whey protein microdroplets, respectively. Directly after preparation, the w/o emulsion was dispensed in sealed 5-g glass tubes and heated at 75 °C for 20 min in a temperature-controlled water bath, in order to heat-gel the microdroplets and form stable microbeads, and then rapidly cooled in ice water to room temperature. A preliminary experiment had been done to determine the appropriate heat-load to be applied to the emulsion systems to ensure complete heat-gelation of the droplets. Several heating temperature and time combinations were applied to the WPI solution of 150 g.kg−1, and the firmness of the gels was measured using penetrometry. The results showed that heat treatment at 75 °C for 20 min was sufficient to gel the whey protein solution at 150 g.kg−1 as showed by a high increase of the gel firmness as compared to the gel produced with the other temperature/time combinations (results not shown). Centrifugation at 5000 x g for 10 min (Heraeus Biofuge Primo R, IMLAB, Lille, France) through an oil-water interface was performed on the heated emulsions (20 g emulsion + 20 g distilled water in a 50-mL tube) in order to collect the microbeads into water. During this centrifugation, the microbeads were transferred into the water phase where they pelleted. The n-dodecane phase and excess water were pipetted out and replaced by water in which the microbeads were redispersed and remained stable. Subsequent centrifugation and redispersion in water were then used to complete washing of the microbeads and replacement of the n-dodecane with distilled water. This step was repeated twice. A final dispersing step was performed using MUF. The protein content of the resulting whey protein microbeads suspended in MUF was determined spectrophotometrically at 280 nm after having dissolved the beads in 8 M urea, 0.05 M citrate tri-sodium, 5 g.L−1 sodium dodecyl sulfate, 0.2 M 2-amino-2-hydroxymethyl-1,3-propanediol (Tris), and 1.54 mg.g−1 dithiothreitol (DTT) at pH 7.4 and using an experimental extinction coefficient value of 1.2244 L.g−1.cm−1 as in Morand et al. (2011b) and buffer as a blank. Whey protein microbead dispersions were eventually standardized at 70 and 90 g.kg−1protein using MUF as the dispersant. Preparation of the microbeads was performed at least twice of each size of sample. At the end of preparation, the microbeads were assumed to contain no more than traces of n-dodecane and PGPR.
2.3 Characterization of the microbeads
2.3.1 Isoelectric pH value
The isoelectric pH (pI) of the whey protein microbeads suspended in MUF was determined as the pH value at which the electrophoretic mobility of the beads was zero. The pH of the dispersion of whey protein microbeads was varied from 3 to 6 using dilution into a 0.1 M citric acid and sodium phosphate buffer (viscosity 1.1454 Pa.s, refractive index 1.337, and dielectric constant 80.4). The electrophoretic mobility was measured at 50 V and at 20 °C on a Zeta sizer nano ZS equipment (Malvern Instruments, Orsay, France) using a laser at 633 nm. The value of the pI was inferred from interpolation of the mobility to 0. Experiment was performed twice for each type of sample.
2.3.2 Optical microscopy
Samples were imaged at three different steps during the formation of whey protein microbeads, namely before and after heat treatment while being dispersed in n-dodecane and after heat treatment and dispersion in MUF. The microbeads were analyzed using an optical microscope (BX51, Olympus, Rungis, France). Samples were diluted 5× by using the appropriate solvent phase (n-dodecane or MUF depending on the step). A drop of diluted sample was transferred on a microscope slide and covered with a coverslip prior to analysis. Lens of ×10 magnification was used. The size (diameter) of each microbead in the image (three images per sample × 2 replications) was scaled manually, and a histogram analysis tool in Microsoft Excel was used to plot the cumulative frequencies on each class of size.
2.4 Confocal laser scanning microscopy of the gel
Samples for confocal images measurement were prepared as described by Andoyo et al. (2014). Briefly, the bead dispersions were equilibrated at 35 °C, labeled using 0.06 μL of rhodamine B isothiocyanate (RITC) per gram of protein and stirred for 15 min prior to the addition of glucono-delta-lactone (GDL, from Merck, Sigma Aldrich). It was previously shown that labeling does not affect the course of acid gelation (Vasbinder et al. 2004; Guyomarc’h et al. 2009). GDL was added as a powder into the sample to induce gelation at 35 °C. Immediately after the addition of GDL, ~60 μL of the labeled sample was deposited on a conclave glass slide, covered with a cover slip, sealed with nail varnish, and incubated at 35 °C until pH 4.5. The rest of the sample was incubated in a water bath, and its pH was measured with a pH meter (Consort C931, Bioblock, France) equipped with Inlab 415 pH electrode (Mettler-Toledo, Viroflay, France). When the sample in the bath reached pH 4.5, the slides were imaged at 543 nm through the cover slip using a TE2000-E Nikon C1i inverted confocal laser scanning microscope (CLSM, Nikon, Champigny-sur-Marne, France). Each image was digitized in gray levels as a 512 × 512 pixel matrix (127.3 × 127.3 μm2). Ten images were analyzed per sample.
2.5 Rheological characterizations of the gels
The dispersed microbeads were equilibrated at 35 °C, then 12.63 or 15.10 g.kg−1 of GDL was added to the 70 or 90 g.kg−1 microbead dispersions, respectively, to achieve a final pH of 4.5 within 6 h. The sample was stirred for 1 min, and the suspension was then transferred to the AR2000 rheometer (TA Instruments, Guyancourt, France) equipped with cone-plate geometry (diameter 6 cm and cone angle 4°). Acid gelation was followed by measuring the elastic modulus, G′ (in Pa), with time and pH, using the oscillatory mode at 35 °C, with a frequency of 1 Hz and 0.1% strain. Low-density paraffin was added around the sample to prevent evaporation. The gelation pH was defined as the moment when G′ > 1 Pa, while the value of G′ at pH 4.5 was considered to be the final G′ at 35 °C (G′max). Two measurements were performed for each independent sample.
3 Results and discussion
3.1 Formation and microstructure of whey protein microbeads
3.2 Acid gelation of whey protein microbeads
Hence, it seemed that small particle size favored the development of more elastic acid gels. This result confirmed previous findings and provided further detail on the effective role of particle size in gel formation, through decoupling size from other factors like composition, nature of the interactions, or shape that were also present in Ju and Kilara (1998), Britten and Giroux (2001), or Torres et al. (2011). van Riemsdijk et al. (2011) studied the formation of whey protein particles by using shear flows to control the protein particle size and found that reducing the size of whey protein microparticles from ~20 to 2.5 μm could increase the final G′ value of acid-induced WPI-locust bean gum co-gels more than 15 times. Smaller size of the whey protein microbeads, as in SM systems, means that the number of particles was larger in the system at the same total protein concentration. Considering homogenous spheres of monodisperse radius, one can calculate that ~90 beads of MM and ~400 beads of SM occupy the volume of 1 LM. van Riemsdijk et al. (2011) suggested that the physical interactions between protein particles could be responsible for the network formation of the whey protein clusters. Furthermore, the formation of whey protein clusters was favored by smaller particle size due to an increased number of particle interactions. However, the increase in G′max could not be linearly correlated with the size (or number) of microbeads in the samples, since the MM and LM gelation curves could hardly be differentiated at either tested concentration. Possibly, a maximum threshold value of particle size may exist that is required for the gel to properly fill the sample volume. When using microparticulated whey proteins in dairy gel applications, it is known that too large particles (typically >1–10 μm) reduce the cohesion and firmness of acid gels (Celigueta Torres et al. 2009). Sağlam et al. (2014b) furthermore showed that inclusion of protein-dense microbeads into heat-set whey protein gels strongly lowered their mechanical elasticity.
3.3 Microstructure of the microbead gel
In conclusion, we found that whey protein microbeads of different sizes could be used as a model way to manipulate properties of acid gels by influencing the connectivity and porosity of the gel. Particle interactions were favored by a smaller size of whey protein microbeads as confirmed by the highest G′max value and smallest pore size of the gels made out of the smallest microbeads.
We wish to thank the Directorate General of Higher Education of the Ministry of Education and Culture of the Republic of Indonesia, for funding through scholarship overseas program.
- Celigueta Torres I, Ipsen R, Knudsen JC, Ostergaard BB (2009) Microparticulated whey protein as fat replacer in yoghurt. In: Celigueta Torres I (ed). ETH Zurich, Switzerland, pp 404–407Google Scholar
- Guyomarc’h F, Jemin M, Le Tilly V, et al (2009) Role of the heat-induced whey protein/κ-casein complexes in the formation of acid milk gels: a kinetic study using rheology and confocal microscopy. In: Famelart MH (ed)Google Scholar
- Janhoj T, Ipsen R (2006) Effect of pre-heat treatment on the functionality of microparticulated whey protein in acid milk gels. Milchwissenschaft 61:131–134Google Scholar
- Tadros TF (2013) Emulsion formation, stability, and rheology. In: Tadros TF (ed) Emulsion formation and stability. Wiley-VCH Verlag GmbH & Co. KGaA, pp 1–75Google Scholar