pubs.acs.org/Langmuir © 2009 American Chemical Society
Comparison of the Orogenic Displacement of Sodium Caseinate with the Caseins from the Air-Water Interface by Nonionic Surfactants N. C. Woodward,* A. P. Gunning, A. R. Mackie, P. J. Wilde, and V. J. Morris Institute of Food Research, Norwich Research Park, Colney, Norwich NR4 7UA, U.K. Received January 20, 2009. Revised Manuscript Received March 4, 2009 Displacement of sodium caseinate from the air-water interface by nonionic surfactants Tween 20 and Tween 60 was observed by atomic force microscopy (AFM). The interfacial structure was sampled by Langmuir-Blodgett deposition onto freshly cleaved mica substrates. Protein displacement occurred through an orogenic mechanism: it involved the nucleation and growth of surfactant domains within the protein network, followed by failure of the protein network. The surface pressure at which failure of the protein network occurred was essentially independent of the type of surfactant. The major component of sodium caseinate is β-casein, and previous studies at the air-water interface have shown that β-casein networks are weak, failing at surface pressures below that observed for sodium caseinate. The other components of sodium caseinate are Rs- and κ-caseins. Studies of the displacement of Rs-caseins from air-water interfaces show that these proteins also form weak networks that fail at surface pressures below that observed for sodium caseinate. However, κ-casein was found to form strong networks that resisted displacement and failed at surface pressures comparable to those observed for sodium caseinate. The AFM images of the displacement suggest that, despite κ-casein being a minor component, it dominates the failure of sodium caseinate networks: Rs-casein and β-casein are preferentially desorbed at lower surface pressures, allowing the residual κ-casein to control the breakdown of the sodium caseinate network at higher surface pressures.
Introduction Food foams are stabilized by surface-active molecules that form structures at the air-water interface.1-3 The two major classes of surface-active agents are proteins and surfactants. Proteins generate foams with long-term stability. The proteins are considered to migrate to the interface, partially unfold, and then link together to form an elastic network. On distortion, the elastic energy stored in the network opposes the deformation of the interface and helps to stabilize the interface. By contrast, surfactants rely on a high degree of mobility to counter deformation of the interface through the Gibbs-Marangoni effect. When mixtures of proteins and surfactants are present, they compete for control of the interface, resulting in the loss of the long-term stability of the foam. The surfactants are generally more surface-active than proteins, and given sufficient time at sufficiently high concentrations, they will eventually displace the protein. Through forming networks at the interface the proteins resist displacement by surfactants. Recent use of atomic force microscopy (AFM) to image interfacial structures has revealed how the proteins resist displacement and has generated a new orogenic model for the displacement of proteins from interfaces by surfactants.4 In the orogenic model, it is the heterogeneity of the protein network that ultimately leads to its destruction. As proteins adsorb at the interface, they partially unfold and interact, and the space available for future adsorption of proteins is reduced. Eventually holes are left in the network that can just accommodate single proteins without unfolding or are too small to allow *Author to whom correspondence should be addressed. Tel/Fax: 44 (0) 1603 251 438. E-mail:
[email protected]. (1) Dickinson, E. Colloids Surf., B 1999, 15, 161. (2) Wilde, P. J. Curr. Opin. Colloid Interface Sci. 2000, 5, 176. (3) Wilde, P. J.; Mackie, A. R.; Husband, F. A.; Gunning, A. P.; Morris, V. J. Adv. Colloid Interface Sci. 2004, 108, 63. (4) Mackie, A. R.; Gunning, A. P.; Wilde, P. J.; Morris, V. J. J. Colloid Interface Sci. 1999, 210, 157.
Langmuir 2009, 25(12), 6739–6744
further adsorption. These proteins are easily displaced, and surfactants can enter the holes vacated by the proteins, or the holes that are too small to allow protein adsorption. These nucleated sites can then grow as more surfactant is adsorbed into these regions. The growth compresses the protein network, leading to the refolding of individual proteins and denser packing of the network of protein molecules. For nonionic surfactants, the nucleated sites grow into larger domains, and this expansion leads to a folding and buckling of the protein network. Eventually, at sufficiently high surface pressures, the network can no longer resist compression, and it breaks. At this point, the protein is present as individual molecules or protein aggregates in a sea of surfactant molecules, and these molecules and aggregates can now be displaced into the bulk. The important finding is that substantial protein displacement cannot occur until the elastic network structure is broken and that this occurs at a characteristic surface pressure determined by the nature of the protein and the adsorption process.4 Displacement by ionic surfactants is also orogenic. However, because of the charge on the surfactant, displacement occurs through the appearance and eventual coalescence of a large number of small domains, rather than the extended growth of a few surfactant domains.5-7 Orogenic displacement has also been observed at both oil-water and air-water interfaces.8-10 For nonionic surfactants, the surfactant domains become large enough to visualize by optical microscopy. Orogenic displacement (5) Gunning, P. A.; Mackie, A. R.; Gunning, A. P.; Woodward, N. C.; Wilde, P. J.; Morris, V. J. Biomacromolecules 2004, 5, 984. (6) Gunning, P. A.; Mackie, A. R.; Gunning, A. P.; Wilde, P. J.; Woodward, N. C.; Morris, V. J. Food Hydrocolloids 2004, 18, 509. (7) Mackie, A. R.; Gunning, A. P.; Wilde, P. J.; Morris, V. J. Langmuir 2000, 16, 8176. (8) Mackie, A. R.; Gunning, A. P.; Wilde, P. J.; Morris, V. J. Langmuir 2000, 16, 2242. (9) Mackie, A. R.; Gunning, A. P.; Wilde, P. J.; Patino, J. M. R. Biomacromolecules 2001, 2, 1001. (10) Morris, V. J.; Gunning, A. P. Soft Matter 2008, 4, 943.
Published on Web 04/06/2009
DOI: 10.1021/la900241q
6739
Article
has been observed directly on interfaces on a Langmuir trough using Brewster angle microscopy (BAM)9,11,12 and in liquid lamellae as models for foams by light microscopy and fluorescence recovery after photobleaching.13 Competition between bovine serum albumin and Tween 20 at an emulsion interface has been studied in situ using front-face fluorescence spectroscopy.14 This method has the advantage of being able to study the displacement in a nondestructive manner but is unfortunately limited to proteins that undergo a measurable shift in their fluorescence emission spectra from the adsorbed to desorbed state, making it unsuitable for the analysis of proteins such as β-lactoglobulin and β-casein.14 Confocal microscopy has enabled in situ visualization of protein-surfactant interactions at an emulsion interface during displacement by surfactant.15 Finally, orogenic expulsion of protein networks from finite-sized oil droplets in a liquid medium has been followed by monitoring the effect on the deformability of the droplets.16 These studies support the contention that the orogenic model is applicable to protein displacement from interfaces in actual foams and emulsions. Initial studies have been made on simple mixtures of proteins and their displacement by surfactants.17,18 For the system studied, the mixed proteins within the interfacial layers were found to be randomly distributed with no evidence of gross phase separation. For simple mixtures of β-casein and β-lactoglobulin, it was found that the protein (β-lactoglobulin), which alone would form the strongest interfacial network, dominated the failure of the mixed network.17 A similar situation was observed for the displacement of whey protein isolate (WPI) from air-water interfaces by Tweens.19 These results suggest that the major component of the isolate, β-lactoglobulin, determines the final failure of the WPI network. However, it would be of interest to understand how a minor component might affect a mixed protein network, particularly if the minor component was capable of forming strong interfacial networks whereas the major components formed weak networks.10 Sodium caseinate is an interesting system for study: it contains Rs1-, Rs2-, β-, and κ-caseins in a ratio of 4:1:4:1, with the majority of the proteins (Rs1-, Rs2-, and β-caseins are ∼85-90%) being random coil proteins and the remainder being the glycoprotein κ-casein.20 Sodium caseinate resembles β-casein and Rs1-casein in being very surface-active and an effective stabilizer of emulsions and foams.20-22 However, unlike the random coil proteins, which show little evidence of the formation of strong interfacial networks, studies carried out at both air-water and oil-water interfaces have shown that sodium caseinate, like (11) Patino, J. M. R.; Sanchez, C. C.; Ni~no, M. R. R.; Fernandez, M. C. J. Colloid Interface Sci. 2001, 242, 141. (12) Patino, J. M. R.; Sanchez, C. C. Biomacromolecules 2004, 5, 2065. (13) Clark, D. C.; Mackie, A. R.; Wilde, P. J.; Wilson, D. R. Faraday Discuss. Chem. Soc. 1994, 98, 253. (14) Rampon, V.; Genot, C.; Riaublanc, A.; Anton, M.; Axelos, M. A.V.; McClements, D. J. J. Agric. Food Chem. 2003, 51, 2482. (15) Kerstens, S.; Murray, B. S.; Dickinson, E. J. Colloid Interface Sci. 2006, 296, 332. (16) Gunning, A. P.; Mackie, A. R.; Wilde, P. J.; Morris, V. Langmuir 2004, 20, 116. (17) Gunning, A. P.; Mackie, A. R.; Kirby, A. R.; Morris, V. J. Langmuir 2001, 17, 2013. (18) Mackie, A. R.; Gunning, A. P.; Ridout, M. J.; Wilde, P. J.; Morris, V. J. Langmuir 2001, 17, 6593. (19) Woodward, N. C.; Wilde, P. J.; Mackie, A. R.; Gunning, A. P.; Gunning, P. A.; Morris, V. J. J. Agric. Food Chem. 2004, 52, 1287. (20) Dickinson, E. J. Dairy Res. 1989, 56, 471. (21) Murphy, J. M.; Fox, P. F. Food Chem. 1991, 39, 211. (22) Klemaszewski, J. L.; Das, K. P.; Kinsella, J. E. J. Food Sci. 1992, 57, 366. (23) Boyd, J. V.; Mitchell, J. R.; Irons, L.; Musselwhite, P. R.; Sherman, P. J. Colloid Interface Sci. 1973, 45, 478. (24) Benjamins, J.; de Feijter, J. A.; Evans, M. T. A.; Graham, D. E.; Phillips, M. C. Faraday Discuss. Chem. Soc. 1975, 59, 218.
6740 DOI: 10.1021/la900241q
Woodward et al.
κ-casein, forms interfacial networks.23-25 It has been suggested that sodium caseinate may share the ability of κ-caseins to form disulfide linkages that would strengthen its interface structure and would also explain the ability of sodium caseinate to behave like other globular proteins in preventing the coalescence of emulsions, unlike the random coil proteins that fail to show such behavior.20-22 The aim of this study has been to use AFM to visualize the displacement of sodium caseinate from the airwater interface by nonionic surfactants. Images obtained from previous work at IFR has shown that, unlike globular proteins such as β-lactoglobulin, β-casein is easily displaced from the air-water interface by surfactants at low surface pressures.4 It would be of interest to see how κ-casein and the Rs-caseins are displaced under the same conditions. This is of importance not only to determine the extent to which the random coil proteins dominate the behavior of sodium caseinate but also to determine the extent to which a minor component such as κ-casein can influence the displacement of a commercial mixture.
Materials and Methods Samples of sodium caseinate (DMV International) and Tween 60 (polyoxyethylene sorbitan monostearate from Quest International) were supplied by Unilever, PLC. Tween 20 (polyoxyethylene sorbitan monolaurate) was obtained as a 10% solution from Pierce (Rockland, Il). Tween 60 solutions were prepared by heating the surfactant above 60 °C, followed by the addition of hot (60 °C) water while stirring. Once the addition of water was complete, the sample was stirred at 60 °C for 30 min, and stirring was continued during cooling to ambient temperature. Samples of β-casein (C-6905, lot 12H9550), Rs-casein (C-6780, 041K7420), and κ-casein (C-0406, 051K7575) were obtained from Sigma (Poole, U.K.). Protein solutions were prepared with surface pure water obtained from an Elga Elgastat UHQ water purification system. Sampling of the air-water interface was carried out using a PTFE Langmuir trough (surface area 0.05 m2, volume 1 dm3) equipped with one fixed and one movable barrier. Protein films were spread from 2 mg mL-1 solutions to give a surface coverage of 3 mg m-2 at the interface. Appropriate volumes of surfactant were added to the subphase to increase the surface pressure, which was monitored by means of a ground glass Wilhelmy plate. At various surface pressures, the interfacial films were transferred onto freshly cleaved mica using the Langmuir-Blodgett technique. Imaging of transferred Langmuir-Blodgett films was carried out using an East Coast Scientific AFM (ECS Ltd., Cambridge, U.K.). Images were obtained in dc (contact) mode using Nanoprobe silicon nitride cantilevers (nominal force constant 0.38 N m-1, Veeco Instruments Ltd.). Measurements were carried out in a liquid cell under redistilled n-butanol (Sigma Chemicals). Protein coverage was determined by measuring the area and height of the protein regions in the AFM images4 at defined surface pressures using Image Pro 4.5 image analysis software (Media Cybernetics Corp.).
Results Interfacial films of sodium caseinate were spread to a surface pressure of ∼12 mN m-1, deposited onto freshly cleaved mica using the Langmuir-Blodgett method, and imaged to ensure that the protein coverage was uniform. Following the addition of surfactant Tween 20 (T20) or Tween 60 (T60) to the subphase, the interfacial film was sampled as a function of increasing surface pressure. Selected AFM images of the interfacial structure are shown in Figure 1. The light areas correspond to regions of the interfacial film occupied by protein. The surfactant molecules (25) Murray, B .S.; Dickinson, E. Food Sci. Technol. Int. 1996, 2, 131.
Langmuir 2009, 25(12), 6739–6744
Woodward et al.
Article
Figure 1. Representative AFM images of the displacement of sodium caseinate from an air-water interface by Tween 20 (T20) and Tween 60 (T60). The protein film was spread to a surface pressure (π) of 12 mN m-1. The data show the structure of the films resulting from the successive adsorption of the surfactants. (a) T20, π = 20.3 mN m-1, image size 12.5 μm. (b) T20, π = 22.3 mN m-1, image size 10 μm. (c) T20, π = 23.7 mN m-1, image size 15 μm. (d) T60, π = 19.3 mN m-1, image size 10 μm. (e) T60, π = 19.8 mN m-1, image size 15 μm. (f) T60, π = 22.0 mN m-1, image size 15 μm.
are removed by butanol, and the dark areas are representative of the regions of the interfacial film originally occupied by the surfactant. The images shown in Figure 1 clearly demonstrate that for both Tween 20 and Tween 60 the area of the interfacial film occupied by protein decreases with increasing surface pressure. For both surfactants, the sequence of images obtained as a function of increasing surface pressure displays the characteristic three stages of the orogenic displacement mechanism. At low surface pressure, in the presence of Tween 20 (20.3 mN m-1, Figure 1a) and Tween 60 (19.3 mN m-1, Figure 1d) the AFM images show the presence of small holes within the sodium caseinate network. This is consistent with the first stage of the orogenic displacement, where the surfactant molecules establish themselves at small defects within the protein network. Upon increasing the surface pressure further (T20, 22.3 mN m-1, Figure 1b; T60, 19.8 mN m-1, Figure 1e), it can be seen that the protein coverage has decreased and that the surfactant domains have grown and become circular in appearance. In the final stages of displacement, close to the collapse point of the film, the surfactant domains are now large and irregular in shape (T20, 23.7 mN m-1, Figure 1c; T60, 22.0 mN m-1, Figure 1f), leaving a fibrous network with a high degree of connectivity, not dissimilar to the types of networks observed previously for β-lactoglobulin4 and WPI.19 From Figure 2, it is clear that the collapse of the sodium caseinate network occurs over the same range of surface pressures for both Tween 20 and Tween 60. Despite the lower surface activity of Tween 60, the displacement of sodium caseinate occurs over the same range of surface pressures, suggesting that the displacement is dominated by the structure of the protein network, not the structure of the nonionic surfactant. Hence, it is possible to examine the effect of the component proteins on the structure and displacement of the sodium caseinate purely by considering displacement data for Tween 20. Images obtained for the displacement of β-casein by Tween 20 are shown in Figure 3. As observed previously,4 the surfactant domains are essentially circular, indicating uniform compression Langmuir 2009, 25(12), 6739–6744
Figure 2. Protein coverage data for the displacement of sodium caseinate films from an air-water interface with nonionic Tween 20 (9) and Tween 60 (0).
of a weak network as the surfactant domains grow in size. The protein coverage data (Figure 4a) show that the network fails at a relatively low surface pressure of ∼23 mN m-1. As will be discussed later it is important to note that these images share similar characteristics to the displacement of the sodium caseinate network by Tween 20 (Figure 1) over a comparable range of surface pressures. The other components of sodium caseinate are Rs- and κ-caseins. Images obtained for the displacement of Rs-casein by Tween 20 (Figure 5) show that Rs-casein also forms weak protein networks at the air-water interface. These images are similar to those seen previously4 and in the present study (Figure 3) for β-casein, with the displacement being complete by a surface pressure of 23 mN m-1. The data also resembles that seen for sodium caseinate over a similar range of surface pressures (Figure 1). Figure 6 shows AFM data for the displacement of κ-casein from an air-water interface by Tween 20. κ-casein was spread to a surface pressure of 6 mN m-1 and compressed to a surface pressure of 14.4 mN m-1. The addition of Tween 20 to the DOI: 10.1021/la900241q
6741
Article
Woodward et al.
Figure 3. AFM images of the displacement of a pure β-casein film from an air-water interface with Tween 20. (a) Image size 1.6 μm, π = 15.9 mN m-1. The arrows show nucleating surfactant domains. (b) Image size 6.4 μm, π = 16.7 mN m-1. (c) Image size 6.4 μm, π = 19.2 mN m-1.
Figure 4. Protein coverage data for the displacement of Rs-casein ()), β-casein (2), and κ-casein (9) with Tween 20.
Figure 5. AFM images of the displacement of a pure RS-casein film from an air-water interface with Tween 20. (a) Image size 10 μm, π = 21.1 mN m-1. (b) Image size 15 μm, π = 21.5 mN m-1. (c) Image size 20 μm, π = 22.3 mN m-1.
Figure 6. AFM images of the displacement of a pure κ-casein film from an air-water interface with Tween 20. (a) Image size 5 μm, π = 14.8 mN m-1. (b) Image size 5 μm, π = 20.6 mN m-1. (c) Image size 12.5 μm, π = 22.9 mN m-1.
subphase was used to induce the displacement of the film (Figure 6a). At surface pressures
