Langmuir 2004, 20, 6093-6095
6093
Adsorbed and Spread β-Casein Monolayers at Oil-Water Interfaces Julia Maldonado-Valderrama, M. Jose´ Ga´lvez-Ruiz, Antonio Martı´n-Rodrı´guez, and Miguel A. Cabrerizo-Vı´lchez* Biocolloids and Fluid Physics Group, Department of Applied Physics, University of Granada, E-18071 Granada, Spain Received January 19, 2004. In Final Form: May 18, 2004 A previous study (Langmuir 2003, 19, 8436) used a Langmuir type pendant drop film balance to form β-casein monolayers at the air-water interface. The present paper reports the application of that technique to the formation of protein monolayers at liquid interfaces. This technique allows a direct comparison between spread and adsorbed β-casein interfacial behaviors that is presented in terms of their π-A isotherms and static elasticity moduli. Π-A isotherms of adsorbed and spread protein have been compared and found to be fairly similar in shape, stability, and also hysteresis phenomena. Examination of the elasticity moduli of both layers shows a similar analogy although slight differences arise and are interpreted in terms of the protein unfolding extent attained by both procedures at the oil interface.
Introduction Proteins constitute a significant group of natural emulsifiers, and many of their functional properties derive from the structure they adapt at interfaces.1-3 The question about protein interfacial conformation has widespread relevance, as it determines stability and formation of foams (air-water interface) and emulsions (oil-water interfaces). The monolayer technique has been revealed as a useful tool in the clarification of this complex process and has been widely used at the air-water interface.4-6 However, the experimental difficulties added by the liquid interface result in a lack of experimental studies on the matter, and the denaturation process undergone by proteins at liquid interfaces remains fairly unclear.6-8 Axisimetric drop shape analysis (ADSA) applied to the pendant drop technique has been widely utilized in the literature in the study of protein adsorption at fluid interfaces.3,9 Moreover, its use has been recently extended to the study of spread lipid layers at liquid interfaces.10,11 A previous study used a constant pressure pendant drop film balance to form β-casein monolayers at the air-water interface12 by means of two different procedures. The main practical advantage of the technique exposed lies in the possibility of obtaining adsorbed protein monolayers likely * Corresponding author. Telephone: 0034 958 243211. Fax: 0034 958 243214. E-mail address:
[email protected]. (1) Dickinson, E. Colloids Surf. 1989, 42, 191. (2) Green, R. J.; Hopkinson, I.; Jones, R. A. L. Langmuir 1999, 15, 5102. (3) Beverung, C. J.; Radke, C. J.; Blanch, H. W. Biophys. Chem. 1999, 81, 59. (4) Mellema, M.; Clark, F. A.; Husband, F. A.; Mackie, A. R. Langmuir 1998, 14, 1753. (5) Kra¨gel, J.; Grigoriev, D. O.; Makievski, A. V.; Miller, R.; Fainerman, V. B.; Wilde, P. J.; Wu¨steck, R. Colloids Surf., B 1999, 12, 391. (6) Murray, B. S.; Nelson, P. V. Langmuir 1996, 12, 5973. (7) Rodrı´guez Patino, J. M.; Rodriguez Nin˜o, M. R. Colloids Surf., B 1999, 15, 235-252. (8) Benjamins, J.; Lucassen-Reynders, E. H. In Proteins at liquid interfaces, 1st ed.; Mobius, D., Miller, R., Eds.; Elsevier: Amdsterdam, 1998; Chapter 9, p 303. (9) Makievski, A. V.; Fainerman, V. B.; Bree, M.; Wu¨steck, R.; Kra¨gel, J.; Miller, R. J. Phys. Chem. B 1998, 102, 417. (10) Kwok, Y.; Vollhardt, D.; Miller, R.; Li, D.; Neumann, A. W. Colloids Surf., A 1994, 88, 51. (11) Li, J.; Miller, R.; Wu¨stneck, R.; Mo¨hwald, H.; Neumann, A. W. Colloids Surf., A 1995, 96, 295.
to be compared to the spread ones. The present paper reports the application of that technique to the formation of protein monolayers at liquid interfaces, and a comparison between spread and adsorbed β-casein layers at the tetradecane-water interface is presented. Comparison is made by means of their π-A isotherms and static elasticity, and this work attempts to look into the denaturation process undergone by proteins at oil-water interfaces and clarify the interfacial behavior of spread and adsorbed protein. Theoretical Basis The interfacial pressure, π, of a protein layer is the difference between the interfacial energy per unit area of the pure liquid-liquid interface, γ0, and the surface energy measured with the layer in place, γ. The response of the adsorbed molecules to compression and expansion of the interface, represented by a change in area and measured by a change in the interfacial pressure, reflects the magnitude of lateral forces existing in the monolayer and can be used to provide structural information.4,6,13 The surface elasticity modulus is defined as the increase in interfacial tension for an increase in area of an interface element:8,14
|| ) |
| ) x (d dγ ln A) T
2 0
+ (ηdω)2
The surface elasticity is a complex quantity with a storage part (elastic) and a loss part (viscous) resulting from a phase difference that occurs between dA and dπ. When there is no exchange of surfactant with the adjoining bulk solution, that is, when ΓA is constant and the time of deformation is very large compared to the time of rearrangements within the layer, is called the static elasticity modulus and it can be deduced from the equilibrium relationship between interfacial pressure and interfacial area: (12) Maldonado-Valderrama, J.; Wege, H. A.; Rodrı´guez-Valverde, M. A.; Ga´lvez-Ruiz, M. J.; Cabrerizo-Vı´lchez, M. A. Langmuir 2003, 19, 8436. (13) Rodrı´guez Nin˜o, M. R.; Carrera Sa´nchez, C.; Rodrı´guez Patino, J. M. Colloids Surf., B 1999, 12, 161. (14) Lucassen-Reynders, E. H. Adv. Colloid Interface Sci. 1970, 2, 347.
10.1021/la0498307 CCC: $27.50 © 2004 American Chemical Society Published on Web 06/25/2004
6094
Langmuir, Vol. 20, No. 15, 2004
0 )
Letters
(d dπ ln A)
Teq
0 provides structural information of the interface and has been widely used in the study of protein monolayers.4,13-15 Experimental Section Materials. Lyophilized, essentially salt free bovine milk β-casein (90+% by electrophoresis) was purchased from SIGMA Chemical Co. It was stored at -18 °C and used without further purification. The oil phase chosen in this study is n-tetradecane (99+%) purchased from Aldrich, purified by chromatography resins Florisil 60-100 mesh and subsequently filtered with 0.2 µm PTFE filters. The aqueous subphase used is a phosphate buffered saline (PBS) of final pH 7.4. Solutions were prepared daily, and 0.054 µS Milli-Q+ purified water was used for buffer preparation and all other purposes. All experiments were performed at T ) 23 °C. Experimental Setup. The present study uses essentially the same experimental protocol as the one used by Maldonado et al.;12 however, a brief description will be presented below. The experiments were performed with a constant surface pressure penetration Langmuir balance based on axisimetric drop shape analysis (ADSA), which is described in detail in ref 16. To obtain the liquid interface, the drop is immersed in a glass cuvette (Hellman) filled with the less dense phase, providing the adequate media to form the interface, that is kept in a thermostatized cell. The solution droplet is formed at the tip of a coaxial double capillary, connected independently to a double microinjector that enables a subphase exchange that substitutes the bulk solution for clean buffer once the desired amount of protein reaches the interface, leaving a protein monolayer at the interface. Spread protein monolayers are obtained by means of a thorough adaptation of Trurnit’s method17 to the requirements of the pendant drop followed by careful immersion of the drop covered by the monolayer into the oil interface. Further details on the procedure can be found in ref 12.
Results and Discussion π-A Isotherms. Spread monolayers were obtained by the deposition of 1 µL of protein solution onto the drop surface by means of mini-glass rods especially designed for this purpose, followed by 1 µL of clean buffer solution so that any remaining protein adhered to the rod was drawn onto the interface.4 In this manner the isotherms were found to be independent of the amount of protein spread on the surface, indicative of reproducible spreading.13 Adsorbed protein monolayers were obtained following the protocol described by Maldonado et al., and the resulting isotherms obtained by the two methods are shown in Figure 1. The compression-expansion speed is a critical parameter in the acquisition of reliable isotherms. Films often appear to have a higher A when they are compressed too quickly.6 In this work, and in both methods, the compression velocity chosen was 0.022 mm2/s, that is, slow enough so that the π-A isotherms obtained represent the equilibrium isotherm. It was observed for both methods that, if a protein film was left for 3 h or more at the maximum area and then recompressed, the π-A curve appeared slightly more expanded on the first compression. If this was immediately followed by second and subsequent compressions, the original π-A curve was recovered. This behavior has also been found for bovine serum albumin (BSA) and β-lactoglobulin and justified with the slow unfolding process undertaken by these proteins at the air interface, which results in a residual expansion that is (15) Cicuta, P.; Hopkinson, I. J. Chem. Phys. 2001, 114, 8659. (16) Cabrerizo-Vı´lchez, M. A.; Wege, H. A.; Holgado-Terriza, J. A. Rev. Sci. Instrum. 1999, 70, 2438. (17) Trurnit, H. J. J. Colloid Interface Sci. 1960, 15, 1.
Figure 1. π-A isotherms for adsorbed β-casein monolayers (open symbols) and spread β-casein monolayers (solid symbols).
lost on subsequent compression cycles but reappears if the film is left for several hours at the maximum area.6 Hence, the interaction between the more hydrophobic segments of β-casein and the oil phase might be the explanation for the similar behavior found for this protein at the tetradecane interface.3,18 The respective isotherms obtained by the two methods are represented versus the area of the drop in Figure 1 (upper axis). This allows a direct comparison of the respective isotherms. Furthermore, the spread protein isotherm is also represented versus the specific area of the protein, knowing the amount of protein spread onto the surface (lower axis), and this might provide an indication of the amount of protein adsorbed at the interface. The great similarity found upon comparison of the curves is noteworthy, and it clearly denotes that both methods lead to protein layers with comparable characteristics. The similar considerable hysteresis phenomena found on both isotherms between the compression and the expansion deserve attention. It may be assumed that the oil phase is a better solvent than air for the more hydrophobic parts of the amino acid residues in the protein.6 Thus, the interaction between protein and tetradecane might result in a partially irreversible unfolded configuration that is not completely recovered upon expansion of the interface. This fact might be accomplished by a further unfolding of the polypeptide chain, resulting in a higher effective area per molecule at the oil-water interface than the one found at the airwater interface.12 These results contrast with those found by Graham and Philips at the decane-water interface.19 Their isotherms, obtained by a similar subphase exchange but in a conventional Langmuir surface balance, appear displaced to lower molecular areas. Differences might be related to the different oil phase used. In this sense, Segumpta et al. have shown differences between the behavior of β-casein at the triolein-water interface and that at the trilinolein-water interface.20 According to them, discrepancies come from the differences in the magnitude of the attractive dispersion interactions between the oil phase and protein molecules. Additionally, other authors have found differences with Graham and (18) Segumpta, T.; Razumovsky, L.; Damodaran, S. Langmuir 1999, 15, 6991. (19) Graham, D. E.; Philips, M. C. J. Colloid Interface Sci. 1978, 70, 403. (20) Segumpta, T.; Damodaran, S. J. Colloid Interface Sci. 1998, 206, 407.
Letters
Figure 2. Static elasticity moduli for adsorbed β-casein monolayers (dashed line) and spread β-casein monolayers (solid line).
Philips results at oil interfaces; Makievski et al. report differences related with β-casein behavior upon adsorption onto the tetradecane-water interface,21 and Murray report differences related with BSA and β-lactoglobulin monolayers at this same interface.6 In any case, and contrary to the findings at the air-water interface, the lack of experimental results on protein monolayers at liquid interfaces causes a very limited knowledege of these systems and further analysis on the matter is needed to clarify the interaction. Static Elasticity Modulus. Figure 2 shows the surface elasticity modulus for the adsorbed and spread monolayers as a function of the interfacial pressure. The modulus is obtained as usual, by fitting the isotherm and directly differentiating the correspondent curve. Although the shapes of the π-A isotherms are almost identical for adsorbed and spread layers, the variation of the elasticity modulus is found to be slightly less similar for the adsorbed and the spread monolayer. Given that the final isotherms are obtained from averaging several measurements, the reproducibility of the small differences found in the curves in Figure 2 was verified, with the error being found to be (0.04 mJ/m2. Similar results relating to differences in rheological properties of protein layers that are not observed in the π-A isotherms have also been found by other authors indicating that elasticity is particularly sensitive to structural changes.5 Let us analyze in detail the curves obtained. The static elasticity modulus for β-casein at oil interfaces shows two peaks for adsorbed and also for spread monolayers. A maximum in is related to a configurational transition (21) Makievski, A. V.; Miller, R.; Fainerman, V. B.; Kra¨gel, J.; Wu¨steck, R. In Food Emulsions and Foams, 1st ed.; Dickinson, E., Rodriguez-Patino, J. M., Eds.; Royal Society of Chemistry: Cambridge, 1999; Chapter 22, p 269.
Langmuir, Vol. 20, No. 15, 2004 6095
in the interfacial structure of the protein.4,13,14 According to Cicuta et al., the first peak would correspond to the tail collapse into the subphase, while the second peak would arise from the collapse of a second region of the molecule into the subphase that may well be the loop region.15 It can be seen that each peak is located in the same interfacial pressure region for both adsorbed and spread monolayers, indicative of a similar energetic cost of the configurational transitions that take place in spread and adsorbed β-casein monolayers, respectively.15 Nevertheless, slight differences appear in the magnitude of both peaks between adsorbed and spread layers. While the first peak appears to some extent higher in magnitude for the adsorbed monolayer, the second appears slightly lower. These differences appear very modest but may suggest that the process of unfolding is rather enhanced by the spreading procedure. As indicated by Cicuta et al., the free energy cost of forcing a loop into the subphase increases with loop length, since the ends of the loops must be constrained at the interface, and this effect is not as significant for tails, since only one end remains at the interface.15 Consistent with these considerations, a slightly more unfolded state may well be achieved by spreading the protein onto the drop surface compared to that of the adsorbed protein that is revealed in the magnitude of the peaks, suggesting the existence of slightly longer loops in the spread monolayer.15 Conclusions An experimental technique, previously presented, that enables a direct comparison of the behavior of adsorbed and spread protein monolayers has been successfully applied to the oil-water interface. A similar interfacial structure adopted by adsorbed and spread protein at the oil interface is found in view of the remarkable coincidence offered by their π-A isotherms. However, evaluation of the static elasticity modulus reveals some minor differences between the interfacial configurations adopted by spread and adsorbed β-casein. While the configurational transitions at both monolayers appear to have the same energetic cost, the differences in magnitude of the peaks between spread and adsorbed β-casein might indicate a slightly further unfolded state of the spread protein that may be induced by the spreading procedure. Acknowledgment. Financial support from “Ministerio de Ciencia y Tecnologı´a, plan nacional de Investigacio´n cientı´fica, Desarrollo e Innovacio´n Tecnolo´gica (I+D+I)”, projects MAT2001-2834-C02-01 and AGL20013483-C02-02, is gratefully acknowledged. Supporting Information Available: Physical properties of the materials used and schematics of the monolayer formation methods. This material is available free of charge via the Internet at http://pubs.acs.org. LA0498307
