Langmuir 2008, 24, 11483-11488
11483
Adsorption and Structural Change of β-Lactoglobulin at the Diacylglycerol-Water Interface Mirian M. Sakuno,*,† Shinya Matsumoto,† Shigeru Kawai,‡ Koseki Taihei,§ and Yasuki Matsumura† Laboratory of Quality Analysis and Assessment, DiVision of Agronomy and Horticultural Science, Graduate School of Agriculture, Kyoto UniVersity, Gokasho, Uji, Kyoto 611-0011, Health Care Food Research Laboratories, KAO Corporation, Sumida-ku, Tokyo 131-8501, and School of Human EViromental Sciences, Mukogawa Women’s UniVersity, Nishinomiya, Hyogo 663-8558, Japan ReceiVed June 11, 2008. ReVised Manuscript ReceiVed August 8, 2008 Diacylglycerol (DAG)/water and triacylglycerol (TAG)/water emulsions were prepared using β-lactoglobulin (βLG) as an emulsifier. The oil phase (20% in emulsion) was mixed with β-LG solution (1% β-LG in water, pH 7) to prepare the emulsions. A fine oil-in-water emulsion was produced from both DAG and TAG oils. The interfacial protein concentration of the TAG emulsion was higher than that of the DAG emulsion. The ζ potential of the DAG oil droplet was higher than that of the TAG oil droplet. The front-surface fluorescence spectroscopy results revealed that tryptophan residues in β-LG moved to the more hydrophobic environment during the adsorption of protein on the oil droplet surfaces. Changes in secondary structure of β-LG during the adsorption were determined by FT-IR spectroscopy. Decreases in the β-sheet content concomitant with increases in the R-helix content were observed during the adsorption to the oil droplets, and the degree of structural change was greater for β-LG in the TAG emulsion than in the DAG emulsion, indicating the increased unfolding of adsorbed β-LG on the TAG oil droplet surface. Results of interfacial tension measurement supported this speculation, that is, the increased unfolding of the protein at the TAG-water interface. Trypsin- and proteinase K-catalyzed proteolysis was used to probe the topography of the adsorbed β-LG on the oil droplet surface. SDS-PAGE analyses of liberated peptides after the proteolysis indicated the higher susceptibility of β-LG adsorbed on the DAG oil droplet surface than on the TAG oil droplet surface. On the basis of all the results, we discussed the conformation of the adsorbed β-LG on the two oil droplet surfaces.
Introduction Obesity is a serious problem for human health particularly in developed countries. Obesity poses a major risk for serious dietrelated chronic diseases, including type 2 diabetes, cardiovascular disease, hypertension and stroke, and certain forms of cancer.1,2 The best way to prevent obesity is by decreasing calorie intake and increasing the opportunity for physical exercise. However, it is difficult for many people to decrease the intake of highcalorie fats, because fats impart the palatability or desired texture to many foods. Fat substitutes are in demand that have similar physical properties but lower calories than normal fats. Diacylglycerol (DAG) attracts interest because of its beneficial effects of decreasing body fat and weight compared with ordinary food oil, that is, triacylglycerol (TAG). Many studies using animals and humans as subjects have proved such effects of DAG.3-10 DAG is a minor component of various edible oils,11 * To whom correspondence should be addressed. E-mail: mirian.sakuno@ gmail.com. † Kyoto University. ‡ KAO Corp. § Mukogawa Women’s University.
(1) Pi-Sunyer, F. X. Am. J. Clin. Nutr. 1991, 53, 1595–1603. (2) World Health Organization (WHO). See the WHO official Web site: www.who.imt/nut/obs.htm. (3) Nagao, T.; Watanabe, H.; Goto, N.; Onizawa, K.; Taguchi, H.; Matsuo, N.; Yasukawa, T.; Tsushima, R.; Shimasaki, H.; and Itakura, H. J. Nutr. 2000, 130, 782–797. (4) Yamamoto, K.; A. H.; Tokunaga, K.; Watanabe, H.; Matsuo, N.; Tokimitsu, I.; Yagi, N. J. Nutr. 2001, 131, 3204–3207. (5) Maki, K. C.; Davidson, M. H.; Tsushima, R.; Matsuo, N.; Tokimitsu, I.; Umporowicz, D. M.; Dicklin, M. R.; Foster, G. S.; Ingram, K. A.; Anderson, B. D.; Frost, S. D.; Bell, M. Am. J. Clin. Nutr. 2002, 76, 1230–1236. (6) Murase, T.; Aoki, M.; Wakisaka, T.; Hase, T.; Tokimitsu, I. J. Lipid Res. 2002, 43, 1312–1319. (7) Tada, N.; Yoshida, H. Curr. Opin. Lipidol. 2003, 14, 29–33. (8) Matsuo, N. Food Sci. Technol. Res. 2004, 10, 103–110.
but the new technology consisting of lipase-catalyzed esterification and refinement enabled the production of food oil containing high DAG contents (more than 80%). Because the physical properties of DAG oil are similar to those of TAG oil, DAG oil has been successfully used for cooking such as frying in Japan. DAG is apt to form water-in-oil (W/O) emulsions. When using low-molecular-weight surfactants for emulsifying DAG oil, the range of oil contents in which oilin-water (O/W) emulsions can be produced is very limited.12 This property of DAG oil enables the production of W/Oemulsion-type foods such as margarine, but is responsible for the difficulty in developing O/W-emulsion-type foods such as dressings, mayonnaise, and creams from DAG oil. In our previous paper, we showed that a DAG emulsion prepared using egg yolk as an emulsifier is less stable than a TAG emulsion, but the use of phospholipase A2-treated egg yolk enables the formation of a fine and stable DAG emulsion.13 Although the experimental data on DAG emulsification using low-molecular-weight surfactants and lipoproteins of egg yolk are available as described above, the emulsification of DAG oil using an ordinary protein (i.e., proteins not conjugated or associated with other compounds such as lipids, carbohydrates, and nucleic acids) has been scarcely reported. Many commercially available food proteins are ordinary proteins. Unless data on the emulsification of DAG oil using ordinary proteins are accumulated, it would be difficult to develop (9) Takase, H.; Shoji, K.; Hase, T; Tokimitsu, I. Atherosclerosis 2005, 180, 197–204. (10) Saito, S.; Tomonobu, K.; Hase, T.; Tokimitsu, I. Nutrition 2006, 22, 417–419. (11) D’alonzo, R. P.; Korazek, W. J.; Wade, R. L. J. Am. Oil Chem. Soc. 1982, 59, 292–295. (12) Shimada, A.; Ohashi, K. Food Sci. Technol. Res. 2003, 9, 142–147. (13) Sakuno, M. M.; Matsumoto, S.; Kawai, S.; Matsumura, Y. Food Hydrocolloids, submitted for publication.
10.1021/la8018277 CCC: $40.75 2008 American Chemical Society Published on Web 09/20/2008
11484 Langmuir, Vol. 24, No. 20, 2008
wide-ranging O/W emulsion foods from DAG oil. The use of whey proteins as natural food ingredients is increasing owing to their excellent surface-active and colloid-stabilizing characteristics.14 The major whey proteins such as β-lactoglobulin (β-LG), R-lactoglobulin, and bovine serum albumin are amphiphilic molecules that have compact globular structures in their native state. The protein molecules can be adsorbed on the oil droplet surface and form a steric layer that protects the droplets against coalescence, thereby providing the physical stability of the emulsions. β-LG represents approximately 50% of the whey proteins and is a small globular water-soluble protein consisting of 162 amino acids. The molecular structure and functional properties of β-LG are the most studied among the whey proteins.15 Particularly, a lot of data on TAG oil emulsions stabilized by β-LG have been accumulated. Therefore, from both the fundamental and practical viewpoints, β-LG is one of the most appropriate proteins for comparative studies on DAG and TAG emulsification. In this context, the emulsion formations of TAG and DAG oils were compared using β-LG as an emulsifier in this study. The adsorption behavior and structure of the adsorbed layer of β-LG were also investigated at the TAG- and DAG-oil interfaces.
Materials and Methods Materials. DAG and TAG oils were supplied by KAO Corp. (Tokyo, Japan). The tri-, di-, and monoacylglyceride contents of DAG were 11.1%, 88.3%, and 0.7%, and those of TAG oil were 97.4%, 2.6%, and 0%, respectively. The fatty acid compositions of TAG and DAG were similar. β-LG (more than 98% purity) was purchased from Sigma Chemical Corp. (St. Louis, MO). The other analytical reagent-grade reagents were purchased from Wako Chemical (Osaka, Japan), Nacalai Tesque, and Sigma Chemical Corp. Interfacial Tension Measurements. The interfacial tensions of TAG or DAG oil-water interfaces were measured by the Wilhelmy plate method, using a FACE surface tensiometer, CBVP-A3 (Kyowa Kaimen Kagaku Co., Tokyo, Japan), at 25 °C for 30 min. The aqueous phase was water or β-LG solution (0.001% β-LG in water, pH adjusted to 7.0). The interfacial pressure (π) was calculated from π ) γ0 - γt, where γ0 is the interfacial tension between TAG or DAG oil and water and γt the interfacial tension in the case where β-LG is added to the aqueous phase. Emulsion Preparation. An emulsion was prepared by mixing 20% TAG or DAG oil with an 80% β-LG solution (1 wt % β-LG in water, pH adjusted to 7.0). Oil and water were emulsified using Physcotron NS-50 equipment (Nichion Ltd., Chiba, Japan) operated at 22 000 rpm for 3 min at room temperature and sonicated using an ultrasonic homogenizer, US-150 (Nissei, Tokyo, Japan), at 100 µA for 2.5 min. Measurements of Particle Size Distribution. The particle size distribution of resultant emulsions was determined using a Horiba LA-500 laser diffraction particle size analyzer (Horiba Ltd., Kyoto, Japan). The mean droplet diameter and specific surface area were derived from the distribution patterns. Determination of the Interfacial Concentration of Protein. Fresh emulsions (1 mL in a 1.5 mL Eppendorf tube) were centrifuged for 30 min at 35000g using a HIMAC centrifuge, CS120 (Hitachi Koki, Co., Ltd., Tokyo, Japan). The protein contents in the supernatants (unadsorbed protein) were determined by the method of Lowry et al.16 The amount of adsorbed protein was calculated by subtracting the unadsorbed amount from the original protein (14) Dickinson, E. J. Dairy Sci. 1997, 80, 2607–2619. (15) Qi, X. L.; Brownlow, S.; Holt, C.; Sellers, P. Biochem. Biophys. 1995, 129, 567–580. (16) Lowry, O. H.; Rosebrough, N. J.; Farr, A. L.; Randall, R. J. J. Biol. Chem. 1951, 193, 265–275.
Sakuno et al. amount in solution. The interfacial concentration of protein (Γ) was expressed as the weight per unit surface area. ζ Potential Measurements. The emulsions were diluted using distilled water (oil content 0.004%) and were then injected into the measurement chamber of an ELS-Z1 particle electrophoresis instrument (Otsuka Co., Tokyo, Japan). The ζ potential was determined at 25 °C. Measurements of Front-Surface Fluorescence Spectroscopy. Front-surface fluorescence spectra were obtained with an F-3000 fluorescence spectrophotometer (Hitachi Co., Tokyo, Japan). Fresh emulsion samples were diluted 2-fold with water and poured into quartz cells of 10 mm optical length. The quartz cell was fixed in a solid sample holder, 650-0161 (Hitachi, Tokyo, Japan), for the measurements of front-surface fluorescence spectra. The emission spectrum was recorded from 300 to 360 nm at 25 °C. The excitation wavelength was 280 nm. Measurements of Fourier Transform Infrared Spectroscopy (FT-IR). The FT-IR transmission spectra of β-LG in solution and emulsions were recorded using an FTIR-480 Plus spectrophotometer (Jasco Co., Tokyo, Japan). To avoid the problem of absorption bands arising from water in the FTIR spectra, which would obscure the amide I bands of the protein, all the samples were prepared with D2O instead of water. The samples were placed between two CaF2 windows. For each spectrum, 50 scans were carried out at 4 cm-1 resolution. For the secondary structure analysis, the deconvolution of each spectrum was performed using Jasco Spectra Manager software (Windows 95/NT version) according to the methods of Fourier selfdeconvolution (FSD) and the finite impulse response operator (FIRO). The spectra were analyzed by second derivatization and Gaussian curve fitting in the amide I region (1600-1700 cm-1). The secondary structural content was calculated from the relative areas of individual assigned bands in the amide I region. Probing Protein Structure at Interfaces. Fresh emulsions (1 mL in a 1.5 mL Eppendorf tube) were centrifuged at 35000g, and the cream and aqueous phases were separated. The cream phases were transferred to another tube and diluted to 1 mL with phosphate buffer (pH 7.0). The resultant emulsions were then added to trypsin (bovine pancreas) or proteinase K (Tritirachium alkaline proteinase) and incubated at 37 °C for 2 h. The ratio of enzyme to substrate protein (w/w) was adjusted to 0.01. After the incubation, the sample emulsions were heated in a water bath at 100 °C to inactivate the enzyme. After being cooled at room temperature, the emulsions were centrifuged at 35000g for 30 min at 25 °C. The separated cream and aqueous phases were mixed with SDS buffer for electrophoresis (0.125 M Tris-HCl buffer (pH 6.8) containing 0.1% bromophenol blue, 20% glycerol, 4% SDS, and 5% mercaptoethanol). The protein concentration in the SDS buffer was adjusted from 1.5 to 2 mg/mL. SDS-PAGE using the tricine buffer system was carried out according to the method of Schagger.17 Protein bands were fixed in a solution containing 40% methanol and 10% acetic acid for 30 min, before being stained with Bio-Safe Coomassie G-250 (BioRad Laboratories) for 1 h. The destaining was carried out using distilled water. Statistical Analyses. Three replicates were used for all the measurements. By using Microsoft Excel version 2000, all the values were averaged, and the means were reported. The standard deviation for three replicates was also reported. Statistical analysis was performed with Ekuseru-Toukei 2006 (Social Survey Research Information Co., Ltd.).
Results and Discussion Time-Dependent Change in Interfacial Tension. The interfacial tensions of DAG- and TAG-water interfaces were measured using the Wilhelmy plate method. Figure 1a shows the time-dependent changes in interfacial tension for 20 min after the contact of the plate to the interface. Because several seconds is required before the contact of the plate to the interface after (17) Scha¨gger, H.; von Jagow, G. Anal. Biochem. 1987, 166, 368–379.
Adsorption of β-LG at the DAG-Water Interface
Langmuir, Vol. 24, No. 20, 2008 11485 Table 1. Mean Diameter, Interfacial Protein Amount (Γ), and ζ Potential of TAG and DAG Emulsionsa averagec b
mean diameter (µm) Γ (mg/m2) ζ potential (mV)
TAG emulsion
DAG emulsion
2.58 ( 0.09 1.27 ( 0.02 -40 ( 0.40
2.34 ( 0.06 0.75 ( 0.02 -46 ( 0.35
a An emulsion was prepared by mixing 20% TAG or DAG oil with 80% β-LG solution (1% β-LG in water, pH 7). b The mean diameter is represented as d3,2. c Each listed value is the average of triplicate sets of measurements ( standard deviations.
Figure 1. Time-dependent changes in interfacial tension and interfacial pressure at oil-water interfaces. The interfacial tension was measured by the Wilhelmy plate method at 25 °C. The pH of the aqueous phase was adjusted to 7. Key: (9) TAG-water interface, (b) DAG-water interface, (0) TAG-water interface (water included 0.001% β-LG), (O) DAG-water interface (water contained 0.001% β-LG). (b) The interfacial pressure was calculated using the data of (a). Key: (9) TAG-water interface, (b) DAG-water interface.
the formation of the interface in our system, the value at 0 min does not indicate the “true” initial value. When the aqueous phase was just distilled water, i.e., without protein, the interfacial tension of the TAG-water interface (9) was almost constant during 20 min, but a slight increase during 20 min was observed in the case of the DAG-water interface (b). The interfacial tension of the TAG-water interface (25.8 mN/m) was higher than that of the DAG-water interface (14.2 mN/m). In comparison with the TAG molecule, the DAG molecule lacks an esterified fatty acyl group and is more hydrophilic, which causes a substantial decrease in the imbalance of molecular forces across an interface between oil and water. That may be the reason for the lower interfacial tension of the DAG-water interface. When β-LG (0.001%) was added to the aqueous phase, the interfacial tension of the TAG-water interface (0) substantially decreased. The value was 15.3 mN/m at 0 min and declined to 11.2 mN/m after 10 min, followed by a gradual decrease. The interfacial tension of the DAG-water interface (O) was also reduced to 9.4 mN/m by the addition of β-LG, but the timedependent change in the interfacial tension was not clearly observed. Figure 1b shows the time-dependent changes in the interfacial pressure at TAG- and DAG-water interfaces. The interfacial pressure at the TAG-water (9) interface started from 10.5 mN/m and increased to 15.6 mN/m after 20 min. On the other hand, the value at 0 min in the case of the DAG-water interface (b) was 4.8 mN/m and gradually increased to 6.2 mN/m. By simplifying the mechanism, the higher interfacial pressure at the TAG-water interface indicates a larger amount of adsorbed β-LG molecules. However, in addition to the adsorbed amount, the unfolding degree is also of importance for the increase in the interfacial pressure; that is, the increased spreading of the polypeptide chain and amino acid residues after the unfolding of β-LG at the TAG-water interface may cause a further increase in the
interfacial pressure. Conversely, the low value of and the small time-dependent increase in the interfacial pressure at the DAG-water interface suggest a low adsorption amount and lower unfolding of β-LG at the interface. Emulsion Formation. Table 1 shows the mean diameter (d3,2) of oil droplets of TAG and DAG emulsions stabilized by β-LG. The mean diameter was approximately 2.5 µm for both emulsions, indicating the formation of fine oil droplets. As described in the Introduction, the DAG emulsion (O/W emulsion) prepared using low-molecular-weight surfactants is unstable and prone to form a W/O emulsion.12,13 However, in our case, a fine emulsion could be prepared from DAG oil using β-LG as an emulsifier, and the emulsion was stable for several days, i.e., no obvious creaming and coalescence (data not shown). This result indicates the possibility of the formation of a stable emulsion when an ordinary protein such as β-LG is used for emulsifying the DAG oil. Because protein molecules cannot be dissolved in the DAG oil phase unlike the low-molecular-weight surfactants, the protein molecules may stay at the interface after the adsorption and stabilize the oil droplets. Amount of Adsorbed Protein. Table 1 gives the interfacial protein concentration (Γ) of β-LG in TAG and DAG emulsions. Because it is known that Γ is generally between 1 and 3 mg/m2 for a monolayer of adsorbed proteins,14 the values (1.28 and 0.75 mg/m2) of TAG and DAG emulsions in Table 1 indicate that β-LG probably forms a monolayer on the oil droplet surfaces. The Γ of TAG was higher than that of DAG. Such a difference in the adsorbed amount between the two oil droplet surfaces can partially explain the large difference in the interfacial pressure between the two interfaces as shown in Figure 1b. The low β-LG surface coverage in DAG emulsion should be due to the higher polarity of DAG oil as compared to TAG oil. Dickinson and Iveson suggested the relation of the oil polarity and the protein surface coverage in emulsions; i.e., they showed lower β-LG surface coverage at the TAG emulsion interface than at the hydrocarbon (n-tetradecane) interface.18 ζ Potential of Oil Droplets Stabilized by β-LG. To provide insight into the charge density on an adsorbed layer at the plane of shear, the ζ potentials of emulsified DAG and TAG oil droplets were determined (Table 1). The ζ potentials of TAG and DAG oil droplets emulsified by β-LG were -39.89 and -46 mV, respectively. The ζ potential of the DAG oil droplets was greater than that of the TAG droplets, which is advantageous for the electrical repulsion of dispersed oil droplets. To confirm that this difference did not originate from the components of the two oils, we compared the ζ potentials of the TAG and DAG oil droplets that were emulsified using the same emulsifier, Tween 80. Both oil droplets had similar values, approximately -30 mV (data not shown). Tween 80 is a nonionic surfactant, and the rather negative ζ potential of the droplets (at very low ionic strength) arises from (18) Dickinson, E.; Iveson, G. Food Hydrocolloids 1993, 6, 533–541.
11486 Langmuir, Vol. 24, No. 20, 2008
Sakuno et al.
Table 2. Wavelength and Intensity at the Maximum (Imax) Point in Fluorescence Emission Spectra of β-LG in Solution and Emulsionsa λmax (nm) Imax (arbitrary unit)
solution
TAG emulsion
DAG emulsion
330.8 121.2
332.6 73.9
334.2 182.1
a Front-fluorescence spectra of β-LG in solution and emulsions were determined using a fluorescence spectrometer. Emission spectra were recorded from 300 to 380 nm with excitation at 280 nm.
the presence of adsorbed counterions (and also, possibly, charged surfactant impurities in the commercial emulsifier). Since no difference was detected between TAG and DAG oil droplets emulsified by the same emulsifier, Tween 80, we can conclude that the difference in the ζ potential between the TAG oil droplets and DAG oil emulsified by β-LG is attributable to the different states of the β-LG adsorbed layer on the TAG and DAG oil droplet surfaces. The Γ of β-LG in the TAG emulsion was higher than that in the DAG emulsion as described above (Table 1). When the adsorbed protein amount is higher, one can expect the plane of shear to be pushed further away from the oil-water interface. This might well lead to the lower absolute value of the measured ζ potential in TAG emulsions. Fluorescence Spectra of β-LG at Oil Droplet Surfaces. The conformations of β-LG dissolved in water and adsorbed on the oil droplet surfaces were investigated by measuring the fluorescence spectra (excitation wavelength 280 nm, emission wavelength 300-380 nm). Fluorescence spectroscopy is a useful technique to follow tertiary structure transitions in proteins because the intrinsic fluorescence of tryptophanyl residues is particularly sensitive to the polarity of microenvironments along the transition.19,20 λmax (the wavelength of the top peak position in the spectrum) and the intensity at λmax are shown in Table 2. The spectrum at approximately 330 nm reflects the environment around the tryptophan residues. Whereas λmax of β-LG in water was 334.2 nm, those of adsorbed β-LG were 332.6 and 330.8 for the TAG and DAG emulsions, respectively. Such “blue shifts” of λmax indicate the movement of tryptophan residues toward the more hydrophobic environment.20,21 It is likely that tryptophan residues in adsorbed β-LG are accessible to the oil phases. It should be noted that the blue shift of λmax for the DAG emulsion was larger than that for the TAG emulsion, suggesting the more hydrophobic environment of the tryptophan of β-LG adsorbed on the DAG oil droplets. β-LG has a hydrophobic pocket which can bind various hydrophobic substances such as retinol, aldehydes, alcohols, and fatty acids,22-24 and tryptophans 19 and 61 are accessible to hydrophobic substances. DAG may interact more readily than TAG with the hydrophobic pocket in β-LG because of its small size, thereby causing the blue shift of the fluorescence spectra. These results are in agreement with those of Liu et al., who reported that the blue shift when β-LG was bound to DMPG at the least indicated that part of the Trp residue transfers into a more hydrophobic environment due to the interaction of β-LG with the acyl chain of phospholipids.20 The intensity at λmax of the adsorbed β-LG in the emulsions was substantially lower than that of β-LG in water. It is not well established how the changes in the fluorescence intensity at λmax are related to the environmental changes around the tryptophan (19) Viseu, M. I.; Carvalho, T. J.; Costa, S. M. B. Biophys. J. 2004, 86, 2392– 2402. (20) Liu, X.; Shang, L.; Jiang, X.; Dong, S.; Wang, E. Biophys. Chem. 2006, 121, 218–223. (21) Castelain, C.; Genot, C. Biochim. Biophys. Acta 1994, 1199, 59–64. (22) Dufour, E.; Haertle, T. Biochim. Biophys. Acta 1991, 1079, 316–320. (23) Wu, S. Y.; Perez, M. D.; Puyols, P.; Sawyer, L. J. Biol. Chem. 1999, 274, 170–174. (24) Wishnia, A.; Pinder, T. W. J. Biochemistry 1996, 5, 1534–1542.
Figure 2. Contents of secondary structural components of β-LG in solution and emulsions. FT-IR spectra of β-LG in solution and emulsions were recorded. In this experiment, to avoid the interference caused by the water signal, D2O was used to solubilize β-LG. The obtained spectra were processed according to the method described in the Materials and Methods to calculate the contents of secondary structures. A χ2 test showed a significant difference (P < 0.01) in each secondary structural content between β-LG in solution and β-LG at the oil interface.
residues,25 but the decreases in intensity of adsorbed β-LG indicate some conformational changes of β-LG during the adsorption on the oil droplet surfaces from the aqueous phase. The lower intensity for the TAG emulsion than for the DAG emulsion suggests the greater extent of unfolding of β-LG on the TAG oil droplet surface, which is consistent with the results concerning the interfacial pressure (Figure 1b). FT-IR Spectra of β-LG on Oil Droplet Surfaces. The FTIR spectra of β-LG in solution (water) and emulsions were determined, and the obtained spectra were processed according to the method described in the Materials and Methods to calculate the content of secondary structures. The results are shown in Figure 2. Only the results on the intramolecular β-sheet (1620-1630cm-1),R-helix(1652-1658cm-1),β-turn(1662-1675 cm-1), and intermolecular β-sheet (1681-1691 cm-1) are shown, because the contents of other secondary structures are not significantly different among the samples. The content of intramolecular β-sheet for β-LG in solution was approximately 45%. β-LG is known to be a β-barrel-type protein and rich in β-sheets. On the basis of X-ray crystallography data,26,27 the β-sheet content of β-LG is approximately 50%, which is consistent with our FT-IR data. On the other hand, the R-helix content of the β-LG solution as shown in Figure 2 is only 8%, which is lower than the R-helix content (10-15%) calculated from the X-ray crystallography data.28 It is known that FT-IR is suitable for the estimation of the β-sheet content, but not the R-helix content.29 The results of protein secondary structures are in agreement with recent work in which the secondary structure composition for native β-LG was determined to be 46% β-sheet and 16% R-helix in CD spectral measurements.30 The results are also in agreement with FT -IR measurements made by Campbell (25) Rampom, V.; Lethuaut, L.; Mouhous-Riou, N.; Genot, C. J. Agric. Food Chem. 2001, 49, 4046–4051. (26) Papiz, M. J.; Saywe, L.; Eliopoulos, E. E.; North, A. C. T.; Findlay, B. C.; Sivaprasadarao, R.; Jones, T. A.; Kraulis, P. J. Nature 1986, 324, 383–385. (27) Monaco, H. L.; Zanotti, G.; Sapadan, P.; Bolognesi, M.; Sawyer, L.; Eliopoulos, E. E. J. Mol. Biol. 1987, 197, 695–706. (28) Brownlow, S; Morais Cabral, J. H.; Cooper, R.; Flower, D. R.; Yewdall, S. J.; Polikarpov, I.; North, A. C.; Sawyer, L. Structure 1997, 5, 481–495. (29) Byler, D. M.; Susi, H. Biopolymers 1986, 25, 469–487. (30) Kim, D. A.; Cornec, M.; Narsimhan, G. J. Colloid Interface Sci. 2005, 285, 100–109.
Adsorption of β-LG at the DAG-Water Interface
et al.31 Substantial changes in the secondary structure of β-LG were induced during the emulsification. Decreases in the intramolecular β-sheet content concomitant with increases in the R-helix content were observed when β-LG was adsorbed on the oil droplet surfaces. Fang and Dalgleish32 and Lefevre and Subirade33 reported the same tendency concerning the structural changes of β-LG induced by the adsorption on the oil-water interfaces. Normally, hydrophobic environments such as lipid membranes and surfactant micelles enhance the propensity for the R-helical formation of polypeptides.34 Carvalho and Costa reported the cationic lipid DDAB has been found to induce the β-R transition in β-LG.19 The transformation from β-sheet to R-helix induced by the adsorption of β-LG on the TAG and DAG oil droplet surfaces, thus, indicates that the polypeptide chain of β-LG was unfolded sufficiently to be in close contact with the hydrophobic environment in the oil phase. There was a significant difference in the secondary structure of β-LG between the TAG and DAG emulsions. The intramolecular β-sheet content of β-LG in the TAG emulsion was less than that in the DAG emulsion. Conversely, the R-helix content was higher in the TAG emulsion than in the DAG emulsion. This result suggests that the conformational change of β-LG was larger when the protein was adsorbed on the TAG oil droplet surface, which is consistent with the results of the interfacial pressure (Figure 1) and fluorescence spectral measurements (Table 2). The β-turn content of β-LG slightly decreased following emulsification. The intermolecular β-sheet content increased when β-LG was adsorbed on the TAG oil droplet surface, but decreased on the DAG oil droplet surface. Lefevre and Subirade34 have pointed out that the intermolecular β-sheet is quite relevant to the formation of the protein film (two-dimensional network structure) adsorbed at the oil-water interface. The higher amount of adsorbed β-LG on the TAG oil droplet surface (Table 1) may increase the probability of molecular-molecular interactions via the intermolecular β-sheet in the two-dimensional protein film. A more cohesive protein film could be formed because of the increased intermolecular β-sheet content at the TAG-water interface, although further studies using a surface rheological technique are required to clarify this point. Proteolysis of Adsorbed β-LG on the Oil Droplet Surfaces. Proteolytic digestion has been used to probe the topography and conformation of adsorbed proteins on the oil droplet surfaces.35 In this study, trypsin- and proteinase K-catalyzed hydrolysis was attempted for β-LG adsorbed on the TAG and DAG oil droplet surfaces, and the resultant peptides were analyzed by SDS-PAGE. The electrophoretic patterns are shown in Figure 3. In the case of the trypsin-catalyzed digestion, β-LG in solution (in water at pH 7) was barely attacked by the enzyme (lane G). A globular protein such as β-LG is insusceptible to proteolysis without denaturation because of its rigid conformation in the native state.35 On the other hand, β-LG adsorbed on the TAG (H) and DAG (I) oil droplet surfaces was susceptible to trypsin attack, and the digested peptides were clearly observed. This indicates that the conformational change of β-LG during the adsorption at the oil droplet surfaces caused the exposure of the target sites to the enzyme in the aqueous phase. Different results were obtained in the case of proteinaise Κ. Unlike trypsin, the substrate specificity of proteinase K is low, (31) Tcholakova, S.; Denkov, N. D.; Sidzhakova, D.; Campbell, B. Langmuir 2006, 22, 6042–6052. (32) Fang, Y.; Dalgleish, D. G. J. Colloid Interface Sci. 1997, 196, 292–298. (33) Lefevre, T.; Subirade, M. Food Hydrocolloids 2001, 15, 365–376. (34) Lefevre, T.; Subirade, M. J. Colloid Interface Sci. 2003, 263, 59–67. (35) Agboola, S. I.; Dalgleish, D. G. J. Agric. Food Chem. 1996, 44, 3631– 3636.
Langmuir, Vol. 24, No. 20, 2008 11487
Figure 3. SDS-PAGE patterns of β-LG after proteolysis in solution and emulsions. β-LG in solution and emulsions (adsorbed molecules on oil droplet surfaces) was subjected to proteolysis by trypsin and proteinase K. The resultant peptides were analyzed by SDS-PAGE using the tricine buffer system. Key: (A) molecular weight markers, (B) β-LG without proteolysis, (C) proteinase K treatment in solution, (D) proteinase K treatment in the TAG emulsion, (E) proteinase K treatment in the DAG emulsion, (F) β-LG without proteolysis, (G) trypsin treatment in solution, (H) trypsin treatment in the TAG emulsion, (I) trypsin treatment in the DAG emulsion.
and the activity of this enzyme is sufficiently high to digest even a globular protein with native conformation. Because of such a characteristic, proteinase K is often used for experiments on the incorporation into or translocation through biomembranes where proteins outside the biomembranes can be detected by easy digestion using proteinase K.36 In Figure 3, unlike in the result using trypsin, β-LG in water (lane C) was completely digested by proteinase K. This indicates that β-LG can be readily attacked by proteinase K when it is surrounded by the enzyme in the aqueous phase even if the protein is in its the native conformation. Therefore, the peptides that were observed in the TAG (D) and DAG (E) emulsions should have originated from the β-LG site protected from proteinase by the oil phase and the close contact between protein molecules on the oil droplet surface. In the case of trypsin-catalyzed digestion, the amount of β-LG that remained intact was obviously higher for the TAG emulsion (H), whereas the amount of smaller peptides was higher for the DAG emulsion (I), indicating the greater susceptibility of β-LG adsorbed at the DAG oil droplet surface. Enhancement of digestion on the DAG oil droplet surface was confirmed by proteinase K treatment; i.e., several peptides ranging from 6000 to 2000 Da were clearly seen in the TAG emulsion, whereas only a few small peptides were observed in the case of the DAG emulsion. It is possible that such a difference in proteolysis between the TAG and DAG emulsions is due to the Γ of β-LG, where the higher density of β-LG on the TAG oil droplet surface may prevent access of trypsin. Adopting this mechanism, Leaver and Dalgleish37 explained their results in which a trypsin-sensitive peptide of β-casein was not readily hydrolyzed at the tetradecane-water interface but was digestible at the soya oil-water interface. The Γ values of β-casein in their study were 1.5 and 2.8 mg/m2 (almost 2 times greater) for the soya oil and tetradecane emulsions, respectively, which are in agreement with our Γ values, namely, 1.28 and 0.75 g/m2 for the TAG and DAG emulsions, respectively, as shown in Table 1. This difference in Γ between the TAG and DAG emulsions therefore could be one reason for the difference in the susceptibility of β-LG on the two oil droplet surfaces. However, other reasons can also be speculated too. The first reason is that the degree of conformational change of adsorbed (36) Ebeling, W.; Hennrich, N.; Klockow, M.; Metz, H.; Orth, H.; Lang, F. Eur. J. Biochem. 1974, 47, 91–97. (37) Leaver, J.; Dalgleish, D. G. J. Colloid Interface Sci. 1991, 149, 49–55.
11488 Langmuir, Vol. 24, No. 20, 2008
β-LG on the oil droplet surfaces is responsible for the sensitivity of the protein to trypsin. That is, the higher susceptibility of β-LG on the DAG oil droplet surface is attributable to the promoted unfolding of the adsorbed protein molecule. The alternative reason is based on the contact degree of the peptide chain to the oil droplet phase. That is, the lower susceptibility of β-LG on the TAG oil droplet surface can be explained by the topography; namely, the larger part of the peptide chain of the protein molecule is embedded in the oil phase and not readily attacked by the enzyme. Because the increased unfolding of β-LG on the TAG droplet surface compared with that on the DAG-water interface was suggested clearly from the FT-IR spectroscopy results (Figure 2), the first reason is more unlikely. If we adopt the second mechanism, the larger part of the peptide chain of β-LG should be embedded in the TAG oil phase, making it inaccessible. It is likely that the increased unfolding enables the protein molecule to be more flexible, which facilitates the rearrangement of the polypeptide chain, resulting in a close contact to the oil phase. Further studies are required to clarify whether the attack sites of β-LG by the enzyme are protected by the oil phase or the close contact of protein molecules. β-LG is well-known as a major allergen of bovine milk.38-40 To inactivate the allergenic property, the digestion of β-LG is essential. However, previous data suggest that β-LG is resistant to protease digestion at a wide pH range even after heating.41 In this study, β-LG adsorbed on the DAG oil droplet surface was found to be digested efficiently (lanes E and I). These results indicate the possibility that emulsion formation using DAG oil is a useful tool for decreasing the allergenic property of proteins in food design.
Conclusions In this study, the emulsion formations of TAG and DAG oils using β-LG as an emulsifier and the characterization of the adsorbed protein layer on the resultant oil droplet surfaces were investigated. Fine emulsions were produced from both DAG and TAG oils, although the previous reports demonstrated that the formation of stable O/W emulsions was difficult when using low-molecular-weight surfactants and egg yolk as emulsifiers.12,13 (38) Wal, J. M. Allergy 1998, 53, 1013–1022. (39) Sharma, S.; Kumar, P.; Beltzel, C.; Singh, T. P. J. Chromatogr., B 2001, 756, 183–187. (40) Breiteneder, H.; Mills, C. E. N. J. Allergy Clin. Immunol. 2005, 115, 14–22. (41) Reddy, M.; Kella, N. K. D.; Kinsella, J. E. J. Agric. Food Chem. 1998, 36, 737–741.
Sakuno et al.
Normally, protein molecules have an amphiphilic nature and, thus, can act as emulsifiers but cannot be dissolved in oil phases because of macromolecular characteristics, unlike low-molecularweight surfactants and phopholipids in egg yolk. Thus, they are possibly suitable for the emulsification of DAG oil. In the next step, the emulsion formation of DAG oil using other food proteins such as caseins, whey proteins, egg white proteins, and soy proteins should be investigated to explore new O/W-emulsiontype foods containing DAG oil. Proteins normally change their conformation during adsorption on the interfaces, and the resultant conformation is quite relevant to the physicochemical properties of emulsions such as the rheology and long-range stability. In the present study, all the interfacial tension, ζ potential, spectroscopy, and proteolysis results suggested the conformational change of β-LG during adsorption on the TAG and DAG oil droplet surfaces. The conformational change occurred at the secondary structure level (as determined by FT-IR spectroscopy) as well as at the tertiary structure level (as determined by fluorescence spectroscopy). With respect to the extent of conformational change or unfolding on the oil droplet surfaces, the results of interfacial tension (interfacial pressure) measurements and FT-IR spectroscopy demonstrated the increased unfolding of β-LG at the TAG-water interface compared to the DAG-water interface, whereas the results of the ζ potential measurements, fluorescence spectroscopy, and proteolysis of adsorbed β-LG by proteases were ambiguous for the determination of the unfolding degree. In addition to the present data concerning β-LG, additional experiments on the topography of other proteins at the TAGand DAG-water interfaces are required to understand the different unfolding mechanisms of protein molecules on the two oil droplet surfaces. Although β-LG is insusceptible to proteolysis, the digestions by trypsin and proteinase K were markedly facilitated by the adsorption of β-LG on the DAG oil droplet surface. The enhanced digestibility of proteins is not only of nutritional benefit but also advantageous for decreasing the allergenic property of proteins as discussed earlier. As described in the Introduction, the DAG oil itself can decrease body fat. Our results, therefore, indicate the possibility that the healthy O/W emulsion with high nutritional value and biological functions can be produced using DAG oil and food proteins. LA8018277
