FRACTAL CHARACTERISTICS OF MICROSTRUCTURE OF BETA-LACTOGLOBULIN PREPARATIONS AND THEIR EMULSIFYING PROPERTIES

Jacek Leman, Tomasz Dołgań, Michał Smoczyński, Zofia Dziuba
Faculty of Food Sciences, University of Warmia and Mazury in Olsztyn, Poland

ABSTRACT

The emulsifying properties of beta-Lg preparations were dependent on the method the protein was isolated from the sweet whey retentate, protein concentration in emulsion and medium pH. The emulsion stabilized with beta-Lg preparation obtained after salting the protein out with 30% solution at pH 2.0 (beta-Lg 1) had oil droplets with smaller volume surface diameter (d_vs) than the emulsions stabilized with the beta-Lg preparation after alfa-lactalbumin isolation from the whey retentate in mild conditions (55 ^oC/30 min, pH 3.9) (beta-Lg 2). The d_vs value for emulsions was found to decline with an increase in the protein concentration from 0.5 to 2.0% and in the pH value from 5.0 to 7.0. Statistically significant (p ≤ 0.05) decline in d_vs value was found for beta-Lg 2 in the whole range of pH and only at pH 5.0 for beta-Lg 1, for which the increase in the concentration above 1.0% was insignificant (p ≤ 0.05) for d_vs at pH 6.0 and 7.0. The increase in the medium acidity from pH 7.0 to 6.0 and 5.0 increased significantly (p ≤ 0.05) d_vs value only for beta-Lg 1 at the concentration 0.5 to 2.0% and for beta-Lg 2 at the concentration 1.0-2.0% and 0.5% , respectively. Natural fractal dimension (D_L) of particles of beta-Lg preparations were almost the same, i.e. 1.31 and 1.32, and their high value reflected developed surface of the protein particles evidenced on electronograms, what was however unimportant to the emulsifying properties.

Key words: beta-lactoglobulin preparations, emulsifying properties, computer image analysis, Sauter’s diameter (d_vs), fractal analysis.

INTRODUCTION

Beta-lactoglobulin (beta-Lg) belongs to natural emulsifiers, i.e. surface active compounds which decrease the surface tension at the o/w interface and promote the formation of emulsion [2]. Stability of the protein film at the o/w interface is determined by diffusion and adsorption of a protein and its structure unfolding enabling rearrangement, reorientation and interaction at the interface [3, 8]. The amount of protein concentration adsorbed on the interface is dependent, among others, on the protein concentration, pH, ionic strength and time of emulsifying [6]. The emulsifying properties of beta-Lg are also notably dependent on the technology used to obtain the protein [9]. Different functional properties of particular whey proteins stimulate a rapid development of their industrial fractionation towards isolation of beta-Lg with native conformation and required functional properties and of alfa-lactalbumin (alfa-La) useful in the technology of foods for infants and babies because of lesser allergenicity than that of beta-Lg.

Characterization of many irregulary-shaped food products by using an electron microscope was up to now limited to a verbal description of structures often difficult for classification and comparison. The concept of fractals proposed by Mandelbrot [11] to the analysis of disordered structures is used more and more frequently in the food science 1, 4, 12, 14, 15, 17] and the determination of ‘natural’ (‘apparent’) fractal dimension of food product particles characterizes well irregularity of their shape and the degree to which their outline or contour is rugged. The functional properties of powdered products (e.g. solubility, behaviour on the oil-water or gas-water interface) are greatly dependent on their surface development and structure. The determination of fractal dimension for the particles of such products opens up the possibility that some functional properties of powdered products will be able to compare.

The use of beta-Lg in the food production demands recognition of the relationship between the structure of the protein and its physico-chemical and functional properties, and of its behaviour in the model systems. The objectives of the present work were to determine: (1) the effect of production method and some environmental factors (pH, protein concentration) on the emulsifying properties of beta-Lg preparations, and (2) the relationship between microstructure of beta-Lg preparations characterized by fractal dimension and their emulsifying properties.

MATERIALS AND METHODS

Material
Beta-Lg preparations used as the material were obtained from a retentate (10 kDa cut-off membrane) of whey after rennet cheese production in a pilot scale through either (a) salting beta-Lg out with 30% NaCl at pH 2.0 according to Mailliart and Ribadeau-Dumas [10] method (preparation beta-Lg 1) or (b) isolation of alfa-La in mild conditions (55^oC/30 min, pH 3.9) according to Pearce (1983) (preparation beta-Lg 2). The protein deposits were lyophilized. The beta-Lg 1 and beta-Lg 2 preparations contained 1.97 and 1.27% of water, respectively, and, on a dry basis 90.34 and 89.63% of total protein, 10.55 and 4.07% of non-protein nitrogen, 1.64 and 2.27% of ash, 0.73 and 0.74% of sodium, 0.08 and 0.37% of calcium, 0.01 and 0.09% of potassium, and 0.008 and 0.054% of magnesium. The beta-Lg preparations were shown to have alfa-La impurity in the amount exceeding 0.2% after examination by polyacrylamide gel electrophoresis.

Scanning electron microscopy
Samples of beta-Lg preparations were applied to carbon discs and then dusted with gold and carbon by sublimation in vacuum using a JEOL JEE 4x apparatus. The samples were viewed in a Tesla BS-300 scanning electron microscope at an accelerating voltage of 15 kV. A series of phothographs of different magnification were obtained which formed a basis for description of the sample microstructures using computer image analysis.

Computer image analysis
The photographs were scanned into a computer. The image was silhouetted at variable grey hue and then its area and perimeter were calculated using MCI Tool software. Because surface (A) ~ perimeter is (P) ^2/DL, were D_L is a fractal dimension of the perimeter, the plot of the function lg (A) = f (lg (P)) is a straight line of the slope a = 2/D_L. This dependency allowed calculation of the natural fractal dimension of the perimeters of the particles.

Emulsifying properties
Soybean oil and buffered solution (50 mmol*dm^-3 phosphate buffer pH 5.0; 50 mmol/dm³ citrate buffer pH 6.0; 50 mmol*dm^-3 imidazole buffer pH 7.0) of beta-Lg were mixed at the oil-water ratio 4:6 (w/w) and the protein concentration 0.5, 1.0 or 2.0% (w/w) in a lab homogenizer MPW 120 (10 000 rev/min, 5 min). The emulsion was deposited without diluting on a micro slide and observed in an optic microscope. The picture from a camera was registered in the memory of an IBM computer with Pentium 200 Processor and Matrox Magic image analysis card. On a basis of oil droplet diameters in emulsion determined with Mocha software (Jandel Scientific, Corte Madera, California, USA), the Sauter diameter (d_vs) was computed from the formula:

(µm)

where: L – number of oil droplets measured in emulsion, d_i – diameter of oil droplet i in emulsion (µm).
The analysis was repeated thrice.

Analytical methods
Beta-Lg preparations were analyzed using standard methods for the content of: dry matter, total nitrogen, 12% TCA soluble nitrogen (Kjeldahl’s method), calcium and magnesium (AAS method after wet mineralization in the mixture of nitric and perchloric acids, 3:1 v/v), sodium and potassium (flame photometric method after the wet mineralization).

Data analysis
Statistical analysis of the results was carried out using Statsoft’s Statistica PL ver. 6.0 software. The significance of differences was determined at p ≤ 0.05.

RESULTS AND DISCUSSION

Microstructure of beta-Lg prepations
Beta-Lg 1 preparation obtained after salting out with 30% NaCl solution at acid medium was heterogenous regarding the particle sizes. The objects sized from 60 to about 620 µm. The microstructure of the preparation was lamellar and the lamella surfaces were flat and smooth (Fig. 1).

Beta-Lg 2 preparation obtained after alfa-La removal from whey retentate in mild conditions was heterogenous regarding not only the particle size, that ranged from about 30 to about 380 µm, but also the structure of particles. Apart from lamellae with developed surface, flat lamellae with regular edges were observed (Fig. 1).

The values for regression coefficients, fractal dimensions and correlation coefficients for logarithmic plots of the surfaces and perimeters of the particles sampled from the beta-Lg preparations are presented in Table 1. The logarithmic dependency of the surface area on the perimeter length is presented in Figure 2.

High values for the determination coefficients for both beta-Lg preparations indicate that the mathematical model used was well filled to the results obtained. Natural fractal dimensions of the particle contours were almost the same for both beta-Lg preparations, i.e. D_L = 1.31 for beta-Lg 1 preparation and D_L = 1.32 for beta-Lg 2 preparation. These data are within the range of values (1.05-1.36) quoted in the literature for particles of powdered food products [15, 16]. High value for the fractal dimension of particle contours for both beta-Lg preparations supported their developed surfaces visible on the electronograms (Fig. 1).

Emulsifying properties
The emulsions obtained were heterogeneous dispersions of spheric oil droplets in water (Fig. 3). The droplets sized 0 to 5 µm, 5 to 10 µm and 10 to 15 µm dominated the freshly made emulsions. Such tendency was observed for the analyzed conditions, irrespective of the type of beta-Lg preparation, protein concentration and pH.

The emulsions stabilized with beta-Lg 1 preparation, obtained after salting out with 30% NaCl solution at pH 2.0 had droplets with smaller volume surface diameter (d_vs) than the emulsions stabilized with beta-Lg 2 preparation, obtained after remowing alfa-La from the whey retentate in mild conditions (Fig. 4). The studies on the effects of protein concentration and medium pH on the d_vs of o/w emulsions stabilized with beta-Lg preparations showed that when the protein concentration and pH value increased, the d_vs values decreased for both beta-Lg preparations. The increase in the protein concentration from 0.5 to 2.0% resulted in statistically significant (p ≤ 0.05) decline in d_vs value in the whole range of pH for beta-Lg 2 preparation and only at pH 5.0 for beta-Lg 1 preparation, for which the increase in the protein concentration above 1.0% was statistically insignificant for d_vs value at pH 6.0 and 7.0 (Fig. 4).

The increase in the medium acidity in the pH range 7.0-6.0 caused statistically significant (p ≤ 0.05) increase in d_vs value for beta-Lg 1 preparation at the protein concentration 0.5 to 2.0% and for beta-Lg 2 preparation at the protein concentration 1.0 and 2.0% . The increase in the medium acidity in the pH range 6.0-5.0 resulted in statistically significant (p ≤ 0.05) increase in d_vs value only for beta-Lg 2 preparation and 0.5% protein concentration, and in statistically significant (p ≤ 0.05) decline in d_vs value for beta-Lg 1 preparation at protein concentration 2.0% (Fig. 4).

The effects of protein concentration (0.5 to 2.0% ) and medium pH (5.0 to 7.0) on the d_vs value were consistent with the studies by other authors [2, 5, 7]. The relationship between the particle surface development of beta-Lg preparations as determined by the value of fractal dimension, and their emulsifying properties was not found. Apart from almost the same values for fractal dimension of particles for both beta-Lg preparations, they differed in the emulsifying properties. It should be then supposed that the structure changes resulting from the fractionation method decide on the behaviour at the interface of protein particles having similar surface development. The fractal dimension reflecting the surface features of protein particles may only by another factor suitable for characterization of protein preparations.

CONCLUSIONS

1. The emulsifying properties of beta-Lg preparations were dependent on the method the protein was isolated from the sweet whey retentate, protein concentration in emulsion and medium pH.

2. The volume surface diameter (d_vs) of oil droplets in emulsion was found to decrease with an increase in the protein concentration from 0.5 to 2.0% and in the pH value from 5.0 to 7.0.

3. The emulsions stabilized with beta-Lg preparation obtained after salting the protein out with 30% NaCl solution at pH 2.0 had oil droplets with smaller d_vs value than the emulsions stabilized with beta-Lg preparation obtained after alfa-La isolation from the whey retentate in mild conditions (55 °C/30 min, pH 3.9).

4. Natural fractal dimension (D_L) of particles of beta-Lg preparations were almost the same, i.e. 1.31 and 1.32, and their high value reflects developed surface of the protein particles evidenced on electronograms, what was however unimportant to the emulsifying properties.

REFERENCES

1. Barret A. H., Peleg M., 1995. Application of fractal analysis to food structure. Lebensm.Wiss. Technol. 28 (6), 553-563.

2. Das K. P., Kinsella J. E., 1989. pH-dependent emulsifying properties of beta-lactoglobulin. J. Dispersion Sci. Technol. 10:77-83.

3. Das K. P., Kinsella J. E., 1990. Stability of food emulsions: physicochemical role of protein and non-protein emulsifiers. Adv. Food Nutr. Res. 34, 144-149.

4. Kalab M., Allan-Wojtas P., Miller Shen S., 1995. Microscopy and other imaging techniques in food structure analysis. Trends Food Sci. Technol. 6, 177-186.

5. Klemaszewski J. L., Haque Z., Kinsella J. E., 1989. An electronic imaging system for determining droplet size and dynamic breakdown of protein stabilized emulsions. J. Food Sci. 54 (2), 440-445.

6. Leman J., Haque Z., Kinsella J. E., 1988. Creaming stability of fluid emulsion containing different milk protein preparations. Milchwissenschaft 50 (12), 666-669.

7. Leman J., Kinsella J. E., 1989. Surface activity, film formation, and emulsifying properties of milk protein. Crit. Rev. Food Sci. Nutr. 28 (2), 115-138.

8. Leman J. 1996. Właœciwoœci emulgujšce beta-laktoglobuliny [Emulsifying properties of beta-lactoglobulin]. Prz. Mlecz., 7:199-202 [in Polish].

9. Leman J. 1998. Wpływ metody otrzymywania preparatów beta-laktoglobuliny na ich skład i właœciwoœci emulgujšce [Influence of production method of beta-lactoglobulin on their composition and emulsifying properties]. Prz. Mlecz., 11, 383-386 [in Polish].

10. Mailliart P., Ribadeau-Dumas B., 1988. Preparation of beta-lactoglobulin and beta-lactoglobulin-free proteins from whey retentate by NaCl salting out at low pH. J. Food Sci., 53 (3), 743-745, 752.

11. Mandelbrot B. B., 1982. The fractal geometry of nature. Freeman WH and Co., New York.

12. Normand M. D., Peleg M., 1986. Determination of the fractal dimension of a particle silhouette using image-processing techniques. Powder Technol. 45, 271-275.

13. Pearce R. J., 1983. Thermal separation of beta-lactoglobulin and alfa-lactalbumin in bovine cheddar cheese whey. J. Dairy Technol., 38, 144-149.

14. Peleg M., 1993. Fractals and foods. Crit. Rev. Food Sci. Nutr. 33 (2), 149-165.

15. Peleg M., Normand M. D., 1985. Characterization of the ruggedness of instant coffee particle shape by natural fractals. J. Food Sci. 50 (3), 829-831.

16. Peleg M., Normand M. D., 1985. Mechanical stability as the limit to the fractal dimension of solid particle silhouettes. Powder Technol. 43, 187-188.

17. Suzuki T., Yano T., 1990. Fractal structure analysis of some food materials. Agric. Biol. Chem. 54, 3131-3135.

Responses to this article, comments are invited and should be submitted within three months of the publication of the article. If accepted for publication, they will be published in the chapter headed 'Discussions' and hyperlinked to the article.