450
Biomacromolecules 2011, 12, 450–456
Particle Diffusion in Globular Protein Gels in Relation to the Gel Structure Gireeshkumar Balakrishnan, Dominique Durand, and Taco Nicolai* Polyme`res, Colloı¨des, Interfaces, UMR CNRS Universite´ du Maine, 72085 Le Mans cedex 9, France Received October 16, 2010; Revised Manuscript Received November 28, 2010
Globular protein gels with a variety of structures were prepared by heating β-lactoglobulin solutions at different concentrations and different ionic strengths. The structure was analyzed in terms of the pair correlation function of the protein concentration, and the volume fraction of the gels was determined. A strong coarsening of the gel structure was observed upon increasing the NaCl concentration between 0.1 and 0.25 M. The mean square displacement of spherical particles with diameters between 0.2 and 2 µm was determined in solutions and in gels by multiparticle tracking of confocal laser scanning microscopy images. Brownian diffusion or trapping of spheres with different sizes was observed, depending on the gel structure. In few cases the diffusion was anomalous. The relationship between gel structure and particle mobility is discussed.
Introduction Globular proteins denature when heated, which in many cases leads to aggregation and, at sufficiently high concentrations, to gelation. Depending on the conditions, transparent homogeneous or opaque heterogeneous gels are formed.1-6 The structure of globular protein gels formed at different pH and ionic strength has been studied in quite some detail using confocal laser scanning microscopy (CLSM) and scattering techniques. It was shown that a dramatic change occurs in the morphology when the strength of the electrostatic interactions between the proteins is varied.7-12 If the electrostatic repulsion is strong, that is, far from the iso-electric point and at low ionic strength, the gels consist of cross-linked strands of protein with a diameter of the same order of magnitude as the proteins themselves. Below a critical value of the pH or above a critical ionic strength, the gels consist of randomly associated spherical particles. It was suggested that the transition is caused by phase separation of the growing aggregates into finite size spherical domains that agglomerate into larger clusters that form a gel above a critical gel concentration or else precipitate.9 Here we explore diffusion of model colloidal particles through well-characterized heat-set gels of β-lactoglobulin (β-lg), which is the main protein component of whey. Our aim is to correlate the structure of gels formed at different ionic strengths with the mobility of particles embedded in the gel. Understanding the mobility of colloidal particles in globular protein gels may also be important for applications. Colloidal particles can be used as vehicles in food products, which can be protein gels, to deliver active ingredients such as drugs, vitamins, or aromas encapsulated in microparticles.13 A large number of studies have already been reported for the self-diffusion of particles in gels and in polymer solutions (see ref 14 for an extensive recent review) and a number of theories have been developed.15,16 However, most of this research is not relevant to particle diffusion through the protein gels that we investigate here. The structure of heterogeneous heat-set protein gels resembles more closely that of randomly aggregated spherical colloids. Recently, we reported on an * To whom correspondence should be addressed. E-mail: taco.nicolai@ univ-lemans.fr.
investigation of the Brownian diffusion of spherical tracers through such gels using numerical simulations.17 The gels were rigid and there was only excluded volume interaction between the gel and the tracer. An important result of that investigation was that the diffusion coefficient in such systems is controlled by the volume fraction that is accessible to the tracers regardless of the gel structure, the gel volume fraction, and the tracer size. A few experiments on tracer diffusion in milk protein gels have been reported in the literature. Cucheval et al.18 attempted to investigate the diffusion of spherical particles in acid milk gels that are formed by coagulation of casein micelles using the same multiparticle tracking technique as is applied in this study. Unfortunately, the fluorescent tracers did not diffuse in the voids, but became stuck in spite of the fact that they were protected by an adsorbed layer of poly(ethylene oxide) (PEO). Mariette and co-workers19-21 were more successful by studying the self-diffusion of PEO chains in these gels using pulsed field gradient NMR. They found that the diffusion of large PEO chains was sensitive to the structure of the gels. Colsenet et al. studied the diffusion of PEO chains in whey protein gels at pH 6.8 and 0.1 M NaCl22,23 and observed a decrease of the diffusion coefficient with increasing protein concentration both before and after gelation. The decrease was more important for larger PEO chains, but depended very little on the ionic strength. In all cases the diffusion was slightly slower in the protein solutions than after gelation. Croguennoc et al. studied the diffusion of dextran in β-lg globulin gels and also found that it was faster than in the corresponding solution before heating.24 Pulsed field gradient NMR can only be used with tracers that are molecules or polymer chains with a diameter smaller than about 0.1 µm. Multiparticle tracking requires larger tracers and is in this sense complementary. We tracked the displacement of fluorescent spheres with diameters between 0.2 and 2 µm in the gels using CLSM. Particle tracking can be used even in highly turbid protein gels over distances up to about 5 µm and has the advantage that the distribution of the mean square displacements can be determined, which potentially gives information about the heterogeneity of the gels. As far as we are aware, the technique has not yet been successfully applied
10.1021/bm101238r 2011 American Chemical Society Published on Web 12/29/2010
Particle Diffusion in Globular Protein Gels
Biomacromolecules, Vol. 12, No. 2, 2011 n
to determine the mobility of tracer colloidal particles in protein gels or in particle gels in general. Here we study diffusion of spherical tracers in β-lg gels at pH 7 formed at different ionic strengths. This system was chosen to explore the possibilities of multiparticle tracking in biopolymer gels, because the structure of β-lg has already been studied in detail and can be varied by fine-tuning the ionic strength. We have studied the effect of the protein concentration, the ionic strength, and the tracer size. A quantitative characterization of the gel structure enabled us to investigate if and to what extent the mobility of colloidal particles is related to the structure. It is perhaps worth stressing that in this study we are looking at diffusion through a rather rigid porous media formed by the protein gel and that we are not probing the elasticity of the gel. The latter can be done if the tracers are trapped in less rigid gels, because their displacement reflects in that case the elasticity of the gel. This method is called microrheology and has been used for instance to study the gelation of β-lg gels at low ionic strength.25
Experimental Section Materials. β-lg was a gift from Lactalis and consisted of a mixture of the variants A and B in about equal quantities. β-lg was dissolved in Milli Q water. The pH of β-lg solutions was set at 7 by adding a small amount of NaOH. The ionic strength was set by adding NaCl. Three mM NaN3 was added as a bacteriostatic. Gels were prepared by heating the β-lg solutions at 80 °C. The heating time was taken sufficiently long so that all the proteins had aggregated and further heating made no difference to the results. A few hours were found to be sufficient in most cases. The tracers were fluorescent spherical latex particles purchased from Polysciences with different diameters: 0.2, 0.5, 1.0, and 2.0 µm. The particles were covered with a layer of adsorbed PEO chains with molar mass 105 g/mol by suspending them in a PEO solution for at least 24 h before use. This was found to be necessary to avoid that the particles stick to the gel. Dynamic light scattering measurements on the smallest particles showed that the PEO layer was 20 nm thick. Measurement after strong dilution in pure water showed that the PEO did not desorb significantly within a few days. A small amount (
