- Email: [email protected]
Interactions, desorption and mixing thermodynamics in mixed monolayers of β-lactoglobulin and bovine serum albumin
Colloids and Surfaces B: Biointerfaces 21 (2001) 19 – 27 www.elsevier.nl/locate/colsurfb
Interactions, desorption and mixing thermodynamics in mixed monolayers of b-lactoglobulin and bovine serum albumin J. Sa´nchez-Gonza´lez a, M.A. Cabrerizo-Vı´lchez b, M.J. Ga´lvez-Ruiz b,* a
Department of Electronics and Electromechanical Engineering, School of Industrial Engineering, Uni6ersity of Extremadura, 06071 Badajoz, Spain b Biocolloid and Fluid Physics Group. Department of Applied Physics, Faculty of Sciences, Uni6ersity of Granada, 18071 Granada, Spain
Abstract The interactions between milk proteins, b-Lactoglobulin (b-Lg) and bovine serum albumin (BSA), at the air– water interface have been evaluated. The surface pressure (y), molecular area (a) isotherms were obtained by compression of the monolayers at different pH and temperature. In the method used to calculate the interactions, the desorbed segments of the proteins into the aqueous subphase have been considered. Earlier, the desorbed segments have been estimated from the compressibility factor, z, as a function of the surface pressure (virial state equation). The main conclusion from this study is that for biopolymers it is not possible to apply only the mixing thermodynamics to evaluate the intermolecular forces. It is necessary to include the desorption phenomenon. From these results, we can conclude that the main interaction between both proteins is of electrostatic character. © 2001 Elsevier Science B.V. All rights reserved. Keywords: Protein mixed monolayers; Interactions; Desorption
1. Introduction The study of the interactions between proteins is of great interest because knowledge can be gained about biological processes with industrial and medical applications. In this case, the research has been carried out using a two dimensional (2-D) molecular arrangement formed by spreaded simple and mixed monolayers of * Corresponding author. Fax: + 34-958-243214. E-mail address: [email protected] (M.J. Ga´lvez-Ruiz).
proteins at the air –liquid interface. By using the monolayer technique, it is possible to get useful molecular information on the behaviour of the proteins and additionally on the mechanisms involved in the competition between two proteins in different processes [1,2]. In this paper, we report on the interactions between two milk proteins, b-Lactoglobulin (bLg) and bovine serum albumin (BSA) at different pH and temperature. According to Costin and Barnes [3], a mixed monolayer shows non-ideal behaviour when its
0927-7765/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved. PII: S 0 9 2 7 - 7 7 6 5 ( 0 1 ) 0 0 1 8 0 - 1
20
properties or simple functions do not depend linearly on the monolayer composition. Following a procedure similar to one chosen by different workers [4– 6], we examine the excess area of mixing, Da E, at various surface pressures as: Da E =NA(a12 −x1a1 −x2a2)
(1)
where NA is the Avogadro number, xk is the mole fraction of the component k, ak is the area of simple monolayer and a12 is the area of mixed monolayers at the same surface pressure used to determine ak. Da E is zero when the films components are immiscible or when they constitute an ideal mixture where particular interactions are absent. Positive values indicate the presence of repulsive interactions between different molecules in mixed monolayers, while attractive interactions cause negative values. In this analysis, it is has been supposed that the whole molecules remain at the interface. However, it is well known that a long-chain, flexible polymer molecule adopts a configuration at the interface in which, depending on the surface pressure, some segments are adsorbed at the interface while others extend as loops into the adjacent liquid phase [5]. To study this phenomenon it has been considered the equilibrium between the desorbed segments, Nd, and the segments that remain at the interface, Ni: z=
&
J. Sa´nchez-Gonza´lez et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 19–27
Ni = e(DGd − ya)/NskT Nd
(2)
where DGd is the potential barrier opposite to the desorption of a whole molecule (it is calculated as follows), Ns is the number of segments in the macromolecule (i.e. amino acid residues for proteins), k is the Boltzmann constant and T the temperature. In this equation, it has been considered that the free energy of desorption per segment is the free energy desorption of a whole molecule per number of segments of the molecule, i.e. it has been assumed that all segments are equivalent in terms of desorption energy.
DGd =
y
a dy
(3)
ye
where ye is the equilibrium surface pressure and y is the surface pressure at which all molecules are completely desorbed. In practice, y has been identified with the maximum surface pressure that the films reach. The fraction of segments desorbed from the interface, r, can be expressed by the relationship [6]:
r= exp
n
DGd − ya +1 NskT
−1
(4)
since r= Nd/(Ni + Nd) or r − 1 = (Ni/Nd)+ 1=z+ 1. The compressibility modulus, K, has been computed from the ya isotherms using: K= − a
(y (a
(5)
T
2. Experimental The experimental device is a Langmuir film balance (Lauda FW-1), which has been described earlier [7,8]. The experimental errors for both area and surface pressure values are smaller than 5%. b-Lg protein was of the highest commercially available purity, obtained from SIGMA (USA), and it is present as dimers. It is supposed that it contains 162×2 amino acid residues and it has a molecular weight of 35 000 g/mol. BSA was supplied with analytical grade by Biokit S.A. (Barcelona, Spain). It is supposed that it contains 582 amino acid residues with a molecular weight of 67 000 g/mol. The liquid subphases were aqueous buffers. The buffers consist of solutions containing sodium phosphate/acetate (pH 3), sodium acetate (pH 4.7) and di-sodium phosphate (pH 7). In all cases, the ionic strength was lower than 1 mM. Water used in all the experiments was first twice distilled in all-Pyrex apparatus and then passed through a Millipore Milli-Q Reagent System for further purification. Conductivity of the water obtained after this process was always 510 − 6/V per cm and its pH was 6–6.5.
J. Sa´ nchez-Gonza´ lez et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 19–27
Fig. 1. Compression isotherms of b-Lg/BSA mixed monolayers at 293.2 K and pH: (a) 3, (b) 4.7 and (c) 7.
21
22
J. Sa´ nchez-Gonza´ lez et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 19–27
Fig. 1. (Continued)
All compounds used in preparing the solutions were analytical quality supplied by Merck. The film-spread process was carried out by using a microsyringe (Hamilton). The protein solutions used were aqueous solutions with the same pH as the subphase in order to avoid potential barriers and to increase the protein spreading. In both cases, a 0.05% (v/v) of amyl alcohol was added to improve the spreading process [8]. Before the compression of the monolayers a 10 min lapse was estimated to be sufficient to obtain protein monolayers at equilibrium. The molecular films were compressed at a constant rate of 6.5× 10 − 4 m/s. Earlier, it was checked that the isotherms are independent of the compression rate in the studied range and not any hysteresis phenomenon has been observed. The isotherms were recorded at (293.29 0.2) K at different pH values and at pH 4.7 at different temperatures.
3. Results and discussion
3.1. Experimental isotherms The compression experimental isotherms of bLg/BSA simple and mixed monolayers at different pHs are shown in Fig. 1. All the isotherms show the typical Sigmoid shape for biopolymers. A decrease in the compressibility of the films during the compression is interpreted as due to the desorption of protein segments from the interface [2,5,8]. The behaviour of simple monolayers as a function of the pH is similar for both proteins. The films are more expanded at the pH (4.7) close to their isoelectric points (i.p.) (5.1 for b-Lg and 4.7 for BSA). This behaviour has been earlier observed for g-Globulin protein [8,9] and it is attributed to a lower solubility of the proteins into the aqueous subphase at this pH. Therefore, the
J. Sa´ nchez-Gonza´ lez et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 19–27
23
Fig. 2. Compression isotherms at pH 4.7 and different temperatures of protein simple monolayers: (a) BSA and (b) b-Lg.
partial desorption of the protein from the interface will be lower too. At pH 4.7 the isotherms corresponding to mixed protein monolayers show a somewhat intermediate behaviour compared with that shown by pure components. This feature indicates small deviations from an ideal mixed systems and consequently, weak intermolecular interactions between b-Lg and BSA. Nevertheless, at pH 3 and 7 the behaviour differs from that observed at pH 4.7. At pH 3 it can be observed that when the concentration of b-Lg is increasing in the mixed monolayers, an important condensing effect takes place. Considering that at this pH both proteins bear a positive net charge, this apparent condensing effect could be due to a desorption process. On the contrary, at pH 7 the mixed monolayers with high content of BSA appear more extended that the BSA simple films. In this case, both proteins bear a negative net charge and the predominant interaction could be repulsive electrostatic. In Fig. 2, the isotherms corresponding to simple monolayers of BSA and b-Lg at pH 4.7 and different temperatures are represented. All the isotherms also show the Sigmoid shape explained in the same terms than before. Both protein films undergo a condensation process as the temperature
increases from 293.2 to 303.2 K. Between 303.2 and 308.2 K an expected expansion of the monolayers is observed. The condensation with an increase in temperature has been observed for other compounds [10,11] and it is interpreted as due to a dehydration process. Then, we can conclude that at 293.2 K the monolayers of BSA and b-Lg are hydrated.
3.2. Miscibility analysis The maximum compressibility modulus (K) and the surface pressures (y) at these maximum values calculated from the experimental isotherms at 293.2 K and different pHs in function of the mixed monolayers composition are included in Table 1. At pH 3 it can be observed Kmax values of the mixed monolayers similar to that found for simple monolayers of b-Lg. These values indicate that Table 1 Maximum values of K (mN/m) in function of the pH and the mole fraction of BSA in the mixed monolayers, at 293.2 K pH/xBSA
0
0.1
0.3
0.5
0.7
0.9
1
3.0 4.7 7.0
47 62 62
51 59 58
50 56 61
48 55 57
48 56 55
45 55 53
33 38 32
24
J. Sa´ nchez-Gonza´ lez et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 19–27
Fig. 3. Miscibility curves (Da E vs. a) for b-Lg/BSA mixed monolayers at different pH: (a) 3, (b) 4.7 and (c) 7.
J. Sa´ nchez-Gonza´ lez et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 19–27
25
Fig. 3. (Continued)
b-Lg could remove BSA protein from the interface bringing about its desorption. At pH 4.7 a clear trend of the Kmax values in function of the composition of the mixed films has not been found. This behaviour confirms the absence of significant interactions between b-Lg and BSA. On the contrary, at pH 7 the mixed monolayers exhibit Kmax values intermediate to those observed by the pure proteins films. In this case, it is revealed the presence of interactions between both proteins. In Fig. 3, the miscibility curves (Da E vs. a) are represented in function of the pH. We observe small deviations from an ideal mixture even at low surface pressures in all range of studied pH. Then, it is possible to conclude that, although the proteins interact between them, the interactions are weak. At pH 3, these apparent interactions are attractive. At pH 4.7, the character of the interactions depends of the surface pressure, since at low surface pressures the deviations are negative and at high surface pressures these are positive. Fi-
nally, at pH 7 the interactions between both proteins are of repulsive character in all range of composition.
3.3. Interactions in protein mixed films It is apparent that a study considering the desorption process is required to explain the interactions and the surface behaviour of the protein mixed monolayers. In Fig. 4, the calculated fraction of the desorbed fragments of both proteins (from the Eq. (4)) in function of the surface pressure, at different pH values, are represented. Comparing both Fig. 3 and Fig. 4, it is possible to elucidate if the observed deviations from an ideal behaviour are due to interactions between proteins or are due to a desorption process. At pH 3, we observe an important desorption effect in all range of compression. Then, the negative deviations from an ideal mixture observed mainly at high proportion of BSA in the mixed films and at low surface pressures, are probably
26
J. Sa´ nchez-Gonza´ lez et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 19–27
due to the desorption of the protein segments. This hypothesis is according to the result of a decrease of the deviations when the surface pressure increases and a readsorption process occurs at the interface. Moreover, at this pH both proteins are positive charged and some repulsive interaction would be observed. If the attractive forces, van der Waals type, are predominant between b-Lg and BSA at this pH, these would be higher at pH 4.7. Nevertheless, at this pH we observe apparent attractive interactions between both proteins at low surface pressures and high concentration of BSA, but the deviations are lower than those found at pH 3 and also a desorption of BSA of ca. 30% takes place at these surface pressures. Moreover, it was expected that these interactions would be again higher when the readsorption takes place, around 20 mN/m, and this fact is not observed. On the contrary, at this surface pressure range some positive deviations are observed. Taking into account that both proteins are uncharged (their i.e.p. are
close to 4.7), the extra interfacial area occupied by the readsorbed segments could be due to that the segments are adsorbed at the interface in a more extended configuration and/or these segments are highly hydrated. The behaviour is completely different at pH 7 than that observed at pH 3 and 4.7. In this case, all the deviations from the ideality are positive indicating repulsive electrostatic interactions between b-Lg and BSA. At this pH both proteins bear a net negative charge. The desorption also is important at this pH. However, this process gives rise to a decrease of the area. Then, it is possible to conclude that in spite of the desorption at pH 7, the main interaction between b-Lg and BSA is repulsive electrostatic. Nevertheless, the desorption could explain the low values of the deviations. Moreover, it is important to take into account that at 293.2 K, the proteins are highly hydrated, this could also explain the low interactions in all studied range of pH.
Fig. 4. Calculated fraction of the desorbed segments assuming that all of them are equivalent in terms of the desorption energy, r, for both proteins, in function of the y, at different pH: (a) 3, (b) 4.7 and (c) 7.
J. Sa´ nchez-Gonza´ lez et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 19–27
The presence of repulsive interactions at pH 7 and not at pH 3 could be due to that at pH 7 the proteins are more charged than at pH 3. Moreover, it is perhaps important to note that above pH 6.5 the b-Lg dimer is known to begin to dissociate enhancing the electrostatic interactions between both proteins.
4. Conclusions After this analysis it is possible to conclude that to study the interactions between biopolymers in mixed monolayers it is necessary to consider the desorption phenomenon during the compression of the films. The application of the mixing thermodynamics alone can induce error mainly when the interactions are of attractive character.
Acknowledgements This work has been supported financially by the Comisio´ n Interministerial de Ciencia y Tecnologı´a (CICYT), projects MAT98-0937-C02-01 and
.
MAT1999-0662-C03-02, by FEDER 1FD97-1366 and by the European INTAS96-1241.
27
project Union,
References [1] K.S. Birdi, Lipid and Bipolymer Monolayers at Liquid Interfaces, Plenum Press, New York, 1989. [2] F. MacRitchie, Chemistry at Interfaces, Academic Press, San Diego, 1990. [3] I.S. Costin, G.T. Barnes, J. Coll. Interf. Sci. 51 (1) (1975) 106. [4] F.C. Goodrich, Proceedings of the 2nd International Congress on Surface Activity, vol. 1, 1957, p. 33. [5] F. MacRitchie, J. Coll. Interf. Sci. 61 (2) (1977) 223. [6] A. Ahluwalia, D. De Rossi, M. Monici, A. Schirone, Biosens. Bielectron. 6 (1991) 133. [7] J. Sa´ nchez-Gonza´ lez, M.A. Cabrerizo-Vı´lchez, M.J. Ga´ lvez-Ruiz, Coll. Polymer Sci. 276 (1998) 239. [8] J. Sa´ nchez-Gonza´ lez, M.A. Cabrerizo-Vı´lchez, M.J. Ga´ lvez-Ruiz, Coll. Surf. B: Biointerf. 12 (1999) 123. [9] M.J. Ga´ lvez-Ruiz, K.S. Birdi, Coll. Polymer Sci. 277 (1999) 1083. [10] A.G. Bois, J. Coll. Interf. Sci. 105 (1985) 124. [11] M.J. Ga´ lvez-Ruiz, M.A. Cabrerizo-Vı´lchez, Coll. Surf. 58 (1991) 61.
Interactions, desorption and mixing thermodynamics in mixed monolayers of b-lactoglobulin and bovine serum albumin J. Sa´nchez-Gonza´lez a, M.A. Cabrerizo-Vı´lchez b, M.J. Ga´lvez-Ruiz b,* a
Department of Electronics and Electromechanical Engineering, School of Industrial Engineering, Uni6ersity of Extremadura, 06071 Badajoz, Spain b Biocolloid and Fluid Physics Group. Department of Applied Physics, Faculty of Sciences, Uni6ersity of Granada, 18071 Granada, Spain
Abstract The interactions between milk proteins, b-Lactoglobulin (b-Lg) and bovine serum albumin (BSA), at the air– water interface have been evaluated. The surface pressure (y), molecular area (a) isotherms were obtained by compression of the monolayers at different pH and temperature. In the method used to calculate the interactions, the desorbed segments of the proteins into the aqueous subphase have been considered. Earlier, the desorbed segments have been estimated from the compressibility factor, z, as a function of the surface pressure (virial state equation). The main conclusion from this study is that for biopolymers it is not possible to apply only the mixing thermodynamics to evaluate the intermolecular forces. It is necessary to include the desorption phenomenon. From these results, we can conclude that the main interaction between both proteins is of electrostatic character. © 2001 Elsevier Science B.V. All rights reserved. Keywords: Protein mixed monolayers; Interactions; Desorption
1. Introduction The study of the interactions between proteins is of great interest because knowledge can be gained about biological processes with industrial and medical applications. In this case, the research has been carried out using a two dimensional (2-D) molecular arrangement formed by spreaded simple and mixed monolayers of * Corresponding author. Fax: + 34-958-243214. E-mail address: [email protected] (M.J. Ga´lvez-Ruiz).
proteins at the air –liquid interface. By using the monolayer technique, it is possible to get useful molecular information on the behaviour of the proteins and additionally on the mechanisms involved in the competition between two proteins in different processes [1,2]. In this paper, we report on the interactions between two milk proteins, b-Lactoglobulin (bLg) and bovine serum albumin (BSA) at different pH and temperature. According to Costin and Barnes [3], a mixed monolayer shows non-ideal behaviour when its
0927-7765/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved. PII: S 0 9 2 7 - 7 7 6 5 ( 0 1 ) 0 0 1 8 0 - 1
20
properties or simple functions do not depend linearly on the monolayer composition. Following a procedure similar to one chosen by different workers [4– 6], we examine the excess area of mixing, Da E, at various surface pressures as: Da E =NA(a12 −x1a1 −x2a2)
(1)
where NA is the Avogadro number, xk is the mole fraction of the component k, ak is the area of simple monolayer and a12 is the area of mixed monolayers at the same surface pressure used to determine ak. Da E is zero when the films components are immiscible or when they constitute an ideal mixture where particular interactions are absent. Positive values indicate the presence of repulsive interactions between different molecules in mixed monolayers, while attractive interactions cause negative values. In this analysis, it is has been supposed that the whole molecules remain at the interface. However, it is well known that a long-chain, flexible polymer molecule adopts a configuration at the interface in which, depending on the surface pressure, some segments are adsorbed at the interface while others extend as loops into the adjacent liquid phase [5]. To study this phenomenon it has been considered the equilibrium between the desorbed segments, Nd, and the segments that remain at the interface, Ni: z=
&
J. Sa´nchez-Gonza´lez et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 19–27
Ni = e(DGd − ya)/NskT Nd
(2)
where DGd is the potential barrier opposite to the desorption of a whole molecule (it is calculated as follows), Ns is the number of segments in the macromolecule (i.e. amino acid residues for proteins), k is the Boltzmann constant and T the temperature. In this equation, it has been considered that the free energy of desorption per segment is the free energy desorption of a whole molecule per number of segments of the molecule, i.e. it has been assumed that all segments are equivalent in terms of desorption energy.
DGd =
y
a dy
(3)
ye
where ye is the equilibrium surface pressure and y is the surface pressure at which all molecules are completely desorbed. In practice, y has been identified with the maximum surface pressure that the films reach. The fraction of segments desorbed from the interface, r, can be expressed by the relationship [6]:
r= exp
n
DGd − ya +1 NskT
−1
(4)
since r= Nd/(Ni + Nd) or r − 1 = (Ni/Nd)+ 1=z+ 1. The compressibility modulus, K, has been computed from the ya isotherms using: K= − a
(y (a
(5)
T
2. Experimental The experimental device is a Langmuir film balance (Lauda FW-1), which has been described earlier [7,8]. The experimental errors for both area and surface pressure values are smaller than 5%. b-Lg protein was of the highest commercially available purity, obtained from SIGMA (USA), and it is present as dimers. It is supposed that it contains 162×2 amino acid residues and it has a molecular weight of 35 000 g/mol. BSA was supplied with analytical grade by Biokit S.A. (Barcelona, Spain). It is supposed that it contains 582 amino acid residues with a molecular weight of 67 000 g/mol. The liquid subphases were aqueous buffers. The buffers consist of solutions containing sodium phosphate/acetate (pH 3), sodium acetate (pH 4.7) and di-sodium phosphate (pH 7). In all cases, the ionic strength was lower than 1 mM. Water used in all the experiments was first twice distilled in all-Pyrex apparatus and then passed through a Millipore Milli-Q Reagent System for further purification. Conductivity of the water obtained after this process was always 510 − 6/V per cm and its pH was 6–6.5.
J. Sa´ nchez-Gonza´ lez et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 19–27
Fig. 1. Compression isotherms of b-Lg/BSA mixed monolayers at 293.2 K and pH: (a) 3, (b) 4.7 and (c) 7.
21
22
J. Sa´ nchez-Gonza´ lez et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 19–27
Fig. 1. (Continued)
All compounds used in preparing the solutions were analytical quality supplied by Merck. The film-spread process was carried out by using a microsyringe (Hamilton). The protein solutions used were aqueous solutions with the same pH as the subphase in order to avoid potential barriers and to increase the protein spreading. In both cases, a 0.05% (v/v) of amyl alcohol was added to improve the spreading process [8]. Before the compression of the monolayers a 10 min lapse was estimated to be sufficient to obtain protein monolayers at equilibrium. The molecular films were compressed at a constant rate of 6.5× 10 − 4 m/s. Earlier, it was checked that the isotherms are independent of the compression rate in the studied range and not any hysteresis phenomenon has been observed. The isotherms were recorded at (293.29 0.2) K at different pH values and at pH 4.7 at different temperatures.
3. Results and discussion
3.1. Experimental isotherms The compression experimental isotherms of bLg/BSA simple and mixed monolayers at different pHs are shown in Fig. 1. All the isotherms show the typical Sigmoid shape for biopolymers. A decrease in the compressibility of the films during the compression is interpreted as due to the desorption of protein segments from the interface [2,5,8]. The behaviour of simple monolayers as a function of the pH is similar for both proteins. The films are more expanded at the pH (4.7) close to their isoelectric points (i.p.) (5.1 for b-Lg and 4.7 for BSA). This behaviour has been earlier observed for g-Globulin protein [8,9] and it is attributed to a lower solubility of the proteins into the aqueous subphase at this pH. Therefore, the
J. Sa´ nchez-Gonza´ lez et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 19–27
23
Fig. 2. Compression isotherms at pH 4.7 and different temperatures of protein simple monolayers: (a) BSA and (b) b-Lg.
partial desorption of the protein from the interface will be lower too. At pH 4.7 the isotherms corresponding to mixed protein monolayers show a somewhat intermediate behaviour compared with that shown by pure components. This feature indicates small deviations from an ideal mixed systems and consequently, weak intermolecular interactions between b-Lg and BSA. Nevertheless, at pH 3 and 7 the behaviour differs from that observed at pH 4.7. At pH 3 it can be observed that when the concentration of b-Lg is increasing in the mixed monolayers, an important condensing effect takes place. Considering that at this pH both proteins bear a positive net charge, this apparent condensing effect could be due to a desorption process. On the contrary, at pH 7 the mixed monolayers with high content of BSA appear more extended that the BSA simple films. In this case, both proteins bear a negative net charge and the predominant interaction could be repulsive electrostatic. In Fig. 2, the isotherms corresponding to simple monolayers of BSA and b-Lg at pH 4.7 and different temperatures are represented. All the isotherms also show the Sigmoid shape explained in the same terms than before. Both protein films undergo a condensation process as the temperature
increases from 293.2 to 303.2 K. Between 303.2 and 308.2 K an expected expansion of the monolayers is observed. The condensation with an increase in temperature has been observed for other compounds [10,11] and it is interpreted as due to a dehydration process. Then, we can conclude that at 293.2 K the monolayers of BSA and b-Lg are hydrated.
3.2. Miscibility analysis The maximum compressibility modulus (K) and the surface pressures (y) at these maximum values calculated from the experimental isotherms at 293.2 K and different pHs in function of the mixed monolayers composition are included in Table 1. At pH 3 it can be observed Kmax values of the mixed monolayers similar to that found for simple monolayers of b-Lg. These values indicate that Table 1 Maximum values of K (mN/m) in function of the pH and the mole fraction of BSA in the mixed monolayers, at 293.2 K pH/xBSA
0
0.1
0.3
0.5
0.7
0.9
1
3.0 4.7 7.0
47 62 62
51 59 58
50 56 61
48 55 57
48 56 55
45 55 53
33 38 32
24
J. Sa´ nchez-Gonza´ lez et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 19–27
Fig. 3. Miscibility curves (Da E vs. a) for b-Lg/BSA mixed monolayers at different pH: (a) 3, (b) 4.7 and (c) 7.
J. Sa´ nchez-Gonza´ lez et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 19–27
25
Fig. 3. (Continued)
b-Lg could remove BSA protein from the interface bringing about its desorption. At pH 4.7 a clear trend of the Kmax values in function of the composition of the mixed films has not been found. This behaviour confirms the absence of significant interactions between b-Lg and BSA. On the contrary, at pH 7 the mixed monolayers exhibit Kmax values intermediate to those observed by the pure proteins films. In this case, it is revealed the presence of interactions between both proteins. In Fig. 3, the miscibility curves (Da E vs. a) are represented in function of the pH. We observe small deviations from an ideal mixture even at low surface pressures in all range of studied pH. Then, it is possible to conclude that, although the proteins interact between them, the interactions are weak. At pH 3, these apparent interactions are attractive. At pH 4.7, the character of the interactions depends of the surface pressure, since at low surface pressures the deviations are negative and at high surface pressures these are positive. Fi-
nally, at pH 7 the interactions between both proteins are of repulsive character in all range of composition.
3.3. Interactions in protein mixed films It is apparent that a study considering the desorption process is required to explain the interactions and the surface behaviour of the protein mixed monolayers. In Fig. 4, the calculated fraction of the desorbed fragments of both proteins (from the Eq. (4)) in function of the surface pressure, at different pH values, are represented. Comparing both Fig. 3 and Fig. 4, it is possible to elucidate if the observed deviations from an ideal behaviour are due to interactions between proteins or are due to a desorption process. At pH 3, we observe an important desorption effect in all range of compression. Then, the negative deviations from an ideal mixture observed mainly at high proportion of BSA in the mixed films and at low surface pressures, are probably
26
J. Sa´ nchez-Gonza´ lez et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 19–27
due to the desorption of the protein segments. This hypothesis is according to the result of a decrease of the deviations when the surface pressure increases and a readsorption process occurs at the interface. Moreover, at this pH both proteins are positive charged and some repulsive interaction would be observed. If the attractive forces, van der Waals type, are predominant between b-Lg and BSA at this pH, these would be higher at pH 4.7. Nevertheless, at this pH we observe apparent attractive interactions between both proteins at low surface pressures and high concentration of BSA, but the deviations are lower than those found at pH 3 and also a desorption of BSA of ca. 30% takes place at these surface pressures. Moreover, it was expected that these interactions would be again higher when the readsorption takes place, around 20 mN/m, and this fact is not observed. On the contrary, at this surface pressure range some positive deviations are observed. Taking into account that both proteins are uncharged (their i.e.p. are
close to 4.7), the extra interfacial area occupied by the readsorbed segments could be due to that the segments are adsorbed at the interface in a more extended configuration and/or these segments are highly hydrated. The behaviour is completely different at pH 7 than that observed at pH 3 and 4.7. In this case, all the deviations from the ideality are positive indicating repulsive electrostatic interactions between b-Lg and BSA. At this pH both proteins bear a net negative charge. The desorption also is important at this pH. However, this process gives rise to a decrease of the area. Then, it is possible to conclude that in spite of the desorption at pH 7, the main interaction between b-Lg and BSA is repulsive electrostatic. Nevertheless, the desorption could explain the low values of the deviations. Moreover, it is important to take into account that at 293.2 K, the proteins are highly hydrated, this could also explain the low interactions in all studied range of pH.
Fig. 4. Calculated fraction of the desorbed segments assuming that all of them are equivalent in terms of the desorption energy, r, for both proteins, in function of the y, at different pH: (a) 3, (b) 4.7 and (c) 7.
J. Sa´ nchez-Gonza´ lez et al. / Colloids and Surfaces B: Biointerfaces 21 (2001) 19–27
The presence of repulsive interactions at pH 7 and not at pH 3 could be due to that at pH 7 the proteins are more charged than at pH 3. Moreover, it is perhaps important to note that above pH 6.5 the b-Lg dimer is known to begin to dissociate enhancing the electrostatic interactions between both proteins.
4. Conclusions After this analysis it is possible to conclude that to study the interactions between biopolymers in mixed monolayers it is necessary to consider the desorption phenomenon during the compression of the films. The application of the mixing thermodynamics alone can induce error mainly when the interactions are of attractive character.
Acknowledgements This work has been supported financially by the Comisio´ n Interministerial de Ciencia y Tecnologı´a (CICYT), projects MAT98-0937-C02-01 and
.
MAT1999-0662-C03-02, by FEDER 1FD97-1366 and by the European INTAS96-1241.
27
project Union,
References [1] K.S. Birdi, Lipid and Bipolymer Monolayers at Liquid Interfaces, Plenum Press, New York, 1989. [2] F. MacRitchie, Chemistry at Interfaces, Academic Press, San Diego, 1990. [3] I.S. Costin, G.T. Barnes, J. Coll. Interf. Sci. 51 (1) (1975) 106. [4] F.C. Goodrich, Proceedings of the 2nd International Congress on Surface Activity, vol. 1, 1957, p. 33. [5] F. MacRitchie, J. Coll. Interf. Sci. 61 (2) (1977) 223. [6] A. Ahluwalia, D. De Rossi, M. Monici, A. Schirone, Biosens. Bielectron. 6 (1991) 133. [7] J. Sa´ nchez-Gonza´ lez, M.A. Cabrerizo-Vı´lchez, M.J. Ga´ lvez-Ruiz, Coll. Polymer Sci. 276 (1998) 239. [8] J. Sa´ nchez-Gonza´ lez, M.A. Cabrerizo-Vı´lchez, M.J. Ga´ lvez-Ruiz, Coll. Surf. B: Biointerf. 12 (1999) 123. [9] M.J. Ga´ lvez-Ruiz, K.S. Birdi, Coll. Polymer Sci. 277 (1999) 1083. [10] A.G. Bois, J. Coll. Interf. Sci. 105 (1985) 124. [11] M.J. Ga´ lvez-Ruiz, M.A. Cabrerizo-Vı´lchez, Coll. Surf. 58 (1991) 61.