From Egg Yolk/κ-Carrageenan dispersions to gel systems: linear viscoelasticity and texture analysis

Accepted Manuscript Title: From Egg Yolk/κ-Carrageenan dispersions to gel systems: Linear Viscoelasticity and Texture Analysis Authors: José Manuel Aguilar, Ana Paula Batista, Ma Cristiana Nunes, Felipe Cordobés, Anabela Raymundo, Antonio Guerrero PII: S0268-005X(10)00175-X DOI: 10.1016/j.foodhyd.2010.08.006 Reference: FOOHYD 1609 To appear in: Food Hydrocolloids Received Date: 31 March 2010 Revised Date: 15 July 2010 Accepted Date: 9 August 2010 Please cite this article as: Aguilar, M., Batista, A.P., Nunes, Cordobés, F., Raymundo, A., Guerrero, A. From Egg Yolk/κ-Carrageenan dispersions to gel systems: Linear Viscoelasticity and Texture Analysis, Food Hydrocolloids (2010), doi: 10.1016/j.foodhyd.2010.08.006 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. From Egg Yolk/κ-Carrageenan dispersions to gel systems: Linear Viscoelasticity and Texture Analysis José Manuel Aguilar1, Ana Paula Batista2, Mª Cristiana Nunes2, Felipe Cordobés1, Anabela Raymundo2 and Antonio Guerrero1, 1 Departamento de Ingeniería Química, Universidad de Sevilla, Facultad de Química, 41012 Sevilla, Spain 2 Nucleo de Investigação em Engenharia Alimentar e Biotecnologia, Instituto Piaget- ISEIT de Almada, Quinta da Arreinela de Cima, 2800-305 Almada, Portugal 10-2 10-1 100101102 10-1 100 101 102 % κ C G' G" 0 0.15 0.30 0.50 G' , G" / (Pa) ω / (rad/s) A From Egg Yolk/κ-Carrageenan dispersions to gel systems: 1 Linear Viscoelasticity and Texture Analysis 2 José Manuel Aguilar1, Ana Paula Batista2, Mª Cristiana Nunes2, Felipe Cordobés1, Anabela 3 Raymundo2 and Antonio Guerrero1, 4 1 Departamento de Ingeniería Química, Universidad de Sevilla, Facultad de Química, 41012 5 Sevilla, Spain 6 2 Nucleo de Investigação em Engenharia Alimentar e Biotecnologia, Instituto Piaget-ISEIT de 7 Almada, Quinta da Arreinela de Cima, 2800-305 Almada, Portugal 8 Abstract 9 The effect of pH (3.5-6) and polysaccharide concentration (0-0.5 wt%) on the linear viscoelastic 10 behaviour of egg yolk/κ-Carrageenan (EY/κC) mixtures in aqueous solution was studied by using 11 small amplitude oscillatory shear (SAOS). Native egg yolk containing 45 wt% solids was used for 12 all the samples. Thermally set EY/κC gels were also studied by SAOS and texture analysis. A 13 variety of linear viscoelastic behaviours depending on κC concentration and pH were exhibited by 14 EY/κC dispersions that may be explained in terms of the contributions of electrostatic attractive 15 interactions and an exclusion volume effect between protein and polysaccharide macromolecules. 16 This last effect seems to be dominant as pH shifts towards the isoelectric point (IEP) of egg yolk 17 proteins, whereas, at pH far from the IEP a certain enhancement in the degree of compatibility, and 18 even some κC autohydrolysis, seems to take place. The results obtained either from rheological or 19 textural characterization of gels were consistent with that balance. In any case, the results obtained 20 suggest that the microstructure of gels is governed by the protein ability to form gels, where 21 hydrophobically driven interactions and subsequent cross linking among protein segments play a 22 dominant role. 23 Keywords: Proteins, Polysaccharides, Gels, Viscoelasticity, Texture. 24 _______________________ 25 A. GUERRERO () 26 Departamento de Ingeniería Química, 27 Universidad de Sevilla, Facultad de Química, 28 41012 Sevilla (Spain) 29 E-mail: aguerr[email protected] 30 Phone: +34 954557179 31 Fax: +34 95455644732 1. Introduction 33 Protein and polysaccharides are natural polymers that are widely used as functional ingredients in 34 food materials. The interactions of these biopolymers, which lead to extremely complicated and 35 interesting phenomena in various systems, impart an important effect on the macroscopic 36 properties of food products such as: flow, stability, texture and mouth feel (de Kruif & Tuinier, 37 2001). Tolstoguzov (2003) stressed out the importance of thermodynamic considerations in 38 protein-polysaccharide interactions to design food formulations with novel or improved properties. 39 Although for very dilute solutions the system is stable since the mixing entropy dominates and 40 proteins and polysaccharides are co-soluble, from a thermodynamic point of view, proteins and 41 polysaccharides may be compatible or incompatible in an aqueous solution. Thermodynamic 42 incompatibility occurs when the interaction between different biopolymers is energetically less 43 favourable than the average interaction between similar biopolymers. This incompatibility, where 44 biopolymers are mutually segregating one from the other, results in the formation of two separated 45 phases, each being rich in one of the biopolymers. If the interaction between both biopolymers is 46 favoured, a phenomenon known as coacervation or associative phase separation of both 47 biopolymers from a phase mainly containing solvent takes place. The main difference between 48 these two mechanisms is that thermodynamic incompatibility is predominantly entropically driven, 49 whereas complex coacervation, which has been attributed to an electrostatic attraction between 50 oppositely charged biopolymers is both entropically and enthalpically driven (Turgeon, Beaulieu, 51 Schmitt & Sánchez, 2003; de Kruif, Weinbreck & de Vries, 2004; Tolstoguzov, 2007). 52 The major non-covalent interactions between proteins and polysaccharides are: electrostatic 53 interactions, steric exclusion, hydrophobic interactions and hydrogen bonding. Thus, associative 54 separation is commonly governed by the electrostatic attraction between molecules with opposite 55 electrical charges, whereas the steric exclusion effect causes segregative separation. The relative 56 importance of these interactions in a particular system depends on the molecules involved, the 57 composition of the aqueous mixture and the environmental conditions. Knowledge of the origin and 58 nature of the interactions involved and the relative importance of these interactions in a particular 59 system can often be used by food scientist to engineer novel structures and physicochemical 60 properties in food systems. Furthermore, by modulating these parameters it is possible to control 61 the interactions between the biopolymers and therefore create microstructures that provide novel 62 textural and sensory properties to food systems, improve emulsion stability to environmental 63 stresses, or enable novel encapsulation-release characteristics (McClements, 2006). 64 In food technology there has been recent increasing interest in protein-polysaccharide complexes 65 and coacervates derived from novel industrial applications such as micro an nano-encapsulation 66 processes, the design of multi-layers structures, the formation and stabilization of food emulsions, 67 the formation of new food gels and the recovery of protein from industrial by-products (Turgeon, 68 Schmitt & Sánchez, 2007). Thus, the rheological and interfacial properties of protein-69 polysaccharide complexes might be advantageously used for designing food products with various 70 textures, stability and delivery properties. In addition, a variety of applications of protein-71 polysaccharide or protein-polyelectrolyte complexes have been described for non-food applications 72 such as microencapsulation for the pharmaceutical industry or its use in the synthesis of 73 biomaterials and the same conclusion might apply to formulate novel delivery systems (Schmitt, 74 2009). 75 The objective of this contribution was to analyse the effect of pH and protein-polysaccharide ratio 76 on the linear viscoelastic behaviour of egg yolk/κ-carrageenan mixtures in aqueous solution and 77 gels obtained after application of thermal treatment to these blends in order to obtain information 78 about the type of microstructures present in the system. Texture properties of gels were also 79 evaluated. 80 2 Materials and methods 81 2.1 Materials 82 Egg Yolk (EY) was obtained from fresh chicken eggs purchased in a local market according to 83 grade A and type L (63-73 g) commercial specifications, and all damaged or cracked eggs were 84 discarded. κ-Carrageenan (κC) (SatiagelTM AMP45) was kindly provided by Degussa (France). 85 The composition for the counter ion of this κC was 1, 14 and 0.4 wt% of sodium, potassium and 86 calcium, respectively. Hydrochloric acid (analytical grade) was obtained from Merck (Germany). 87 2.2 Methods 88 Chicken eggs were hand broken and the white carefully separated from the yolk using the Harrison 89 and Cunningham (1986) preparation method. EY/ κC mixtures were made by dispersing different 90 amounts of κC (0-0.5 wt%) in native EY (50.43 ± 0.8 average solid content wt %), under 91 mechanical stirring (300 rpm, 1h) at room temperature. The natural pH of the system was modified 92 using 1M and 2M hydrochloric acid as acidulate and demineralised water was adding until 45 wt% 93 in solids EY content. 94 Small Amplitude Oscillatory Shear (SAOS) measurements of egg yolk dispersions containing 45 95 wt% in solids at different pH values an κC content, including either stress sweep or frequency 96 sweep, were performed by means of a controlled-stress rheometer Haake RS-300 (Thermo 97 Scientific, Germany), using 60 mm and 4º cone-plate geometry with 1 mm gap. A shear stress 98 sweep test was always done beforehand at 6.3 rad/s to establish the linear viscoelasticity range. 99 All the dynamic viscoelasticity frequency sweep measurements were carried out at 20ºC and at a 100 stress clearly lower than the critical value for linear viscoelasticity. 101 Measurements of the zeta potential of systems containing 45 wt % in EY and 0.3 wt % in κC at 102 different pH values were performed using a Zetasizer 2000 (Malvern Instruments, U.K.). Aliquots 103 of 0.5 g were diluted in deionised water up to 100 ml of total volume, keeping adjusted the pH 104 value of the dispersion with 0.1 N HCl. Prior to analysis, the samples were tempered at 20ºC and 105 stirred in order to keep the suspension homogeneous, avoiding segregation. The samples were 106 measured in triplicate and each measurement was an average of five determinations. During the 107 experiments, no flocculation was observed and each measurement was conducted at 20 °C. The zeta 108 potential was calculated from the electrophoretic mobility using the Henry equation and the 109 Smoluchowski approximation. 110 The gels were obtained by transferring the above samples without entrapping air into a 20 ml 111 plastic bottles (Ø= 20mm, 65mm height) and sealed prior to heat at 90ºC in a thermally controlled 112 water bath during 30 min. These gels were stored overnight at 5ºC in order to allow the maturation 113 of mixed gels. Dynamic linear viscoelastic properties were conducted by frequency sweep tests on 114 thin slices (1.5±0.3mm) of gel samples at 20ºC, using serrated plate-plate geometry with 1 mm gap 115 within the linear response regime. The pH values were also checked after processing and no 116 modification was detected. 117 Texture properties of the final gels were evaluated using a TA-XT2i (Stable Micro Systems, UK) 118 texturometer. In order to avoid any correction for non-ideal diameter/height ratios, the gel samples 119 were cut into 20mm diameter and 26 mm height specimens (Kuhn, 2000).and lubricated with low 120 viscosity silicon oil. Uniaxial compression tests were performed using a 75mm diameter flat platen 121 (15% strain, 600 s and 0.1 mm/s of crosshead speed). Before performing any measurements, gels 122 were allowed to equilibrate at 20ºC for approximately 3 h in a temperature-controlled room. 123 Textural results for each gel were performed at least three times. 124 3. Results and discussion 125 3.1. Linear viscoelastic properties of EY/ κ C dispersions 126 Figure 1 displays the influence of both linear viscoelastic moduli (G’ and G’’) on frequency for EY 127 dispersions as a function of κC concentration at pH 6 (Fig. 1A) and pH 3.5 (Fig. 1B). As may be 128 observed in both graphs, in absence of polysaccharide, the EY dispersion showed a predominantly 129 viscous fluid-like behaviour, with both moduli showing higher values for the lower pH, due to the 130 electrostatic interactions among charged protein surfaces. The addition of κC to EY leaded to a 131 remarkable evolution in both viscoelastic funtions that depends on pH. 132 Near the mean isoelectric point (IEP) of yolk proteins (pH 6), when 0.15 wt% κC was added, a 133 crossover between G’’ and G’ took place at high frequency. Although a further increase in κC up to 134 0.3 wt% yielded similar values of both moduli at high frequency, the behaviour is rather different. 135 Thus, at low frequency G’’ was always higher than G’ for the 0.15 wt% κC/EY, whereas, the blends 136 containing 0.30 and 0.50 wt% κC were predominantly elastic, giving rise to a crossover point and a 137 shift to a predominantly viscous response at higher frequencies. This crossover took place at lower 138 frequency for the highest κC content and may be related to the increase in repulsive interactions 139 due to increasing negative charges. Thus, the zeta potential value at pH 6 increase from almost 0, 140 corresponding to the IEP in absence of κC, to -26.9 mV for 0.3 wt% κC (see Table 1). In any case, 141 no evolution to gel-like behaviour was found near the IEP. This behaviour is similar to that one 142 shown by amorphous uncrosslinked polymers of high molecular weight as described by Ferry 143 (1980). In accordance, the first above mentioned crossover point corresponds to the so-called 144 terminal transition of the generalized mechanical spectrum, whereas the second one corresponds 145 to the beginning of the transition region, with G’’ dominating over G’. For uncrosslinked polymers of 146 high molecular weight, this region is preceded by a predominantly elastic entangled plateau region 147 that in this case was observed at the low frequency regime. 148 At pH 3.5 protein surfaces exhibit a positive net charge, as may be observed in Table 1. The 149 addition of κC leaded to an increase in G’ and G’’, as well as to a decrease in their respective 150 frequency slopes and loss tangent and a reduction in the zeta potential value. However since the 151 amount of protein is much higher than the polysaccharide content the electrostatic repulsions are 152 still noticeable. The EY dispersion containing 0.30% κC experienced a marked decrease in loss 153 tangent and frequency dependence, showing a weak gel-like behaviour, with G’ higher than G’’ for 154 the whole frequency range studied, which indicates that the presence of charges on protein 155 surfaces promoted the enhancement of network-like structures. A further increase in κC up to 156 0.50% resulted in a weakening of the dispersion leading to linear viscoelastic functions similar to 157 that one shown by the system containing 0.15% κC. In fact, occurrence of maximum consistency 158 for mixed biopolymers gels has been described in the literature of food hydrocolloids (Turquois, 159 Rochas & Taravel, 1992). Imeson (2009) has recently reviewed the interactions between κC and 160 other biopolymers, reporting a positive synergism with LBG or konjac glucomannan which yields 161 maximum gel strength at intermediate κC content for 1% total polysaccharide. A synergism 162 between κC and κ-casein protein was also reported, being attributed to interaction of an 163 electrostatic nature with positively charged amino acids at the surface of κ-casein micelles (Hood 164 and Allen, 1977; Langendorff, Cuvelier, Michon, Launay, Parker & de Kruif, 2000; Spagnuolo, 165 Dagleish, Goff & Morris, 2005, Tijssen, Canabady-Rochelle & Mellema, 2007). 166 Figure 2 shows the evolution of tan δ at 1 rad/s as a function of κC content for pH 3.5. Previous 167 results on EY dispersions at acidic pH (3 and 3.5) showed that the behaviour in absence of 168 polysaccharide is quite different from that obtained for 0.3 κC/EY blends, which corresponds to a 169 viscoelastic fluid showing the terminal transition crossover point at high frequency (Guerrero, 170 Carmona, Martínez & Partal, 2004; Aguilar, Cordobés & Guerrero 2007). The results obtained at 171 pH 3.5 suggest that protein-polysaccharide interactions of electrostatic nature may lead to the 172 formation of gel-like network structures, displaying maximum strength at 0.3 wt% κC (minimum 173 loss tangent), at which protein to polysaccharide optimum ratio seems to be found (Schmitt et al. 174 2009). A possible explanation for this increase in the loss tangent above 0.3% is a phenomenon 175 known as carrageenan autohydrolysis, which occurs at low pH values as carrageenan in the acid 176 form cleaves at the 3,6-anhydrogalactose linkage in the molecule (Hoffmann, Russell & Gidley, 177 1996). Although this phenomenon is favoured at high temperature, autohydrolysis would also take 178 place to some extent at temperatures above the cold-gelling point as long as κC content is high 179 enough (i.e. 0.5 % κC). The contribution of the decrease in electrostatic repulsions, although 180 moderate, may also be taken into account, being a consequence of the addition of negatively 181 charged κC. 182 3.2. Linear viscoelastic properties of EY/ κ C gels 183 The mechanical spectra of thermally processed EY/κC systems at pH 6 are shown in Figure 3. As 184 may be observed, these systems exhibit a gel behaviour with G’ being much higher than G’’. The 185 addition of κC leaded to a clear increase in both moduli, but did not produce any modification on 186 the shape of the mechanical spectra, in spite of the variety of viscoelastic behaviours exhibited by 187 the unheated EY/κC mixtures shown in Fig 1. These results indicate that the viscoelastic behaviour 188