928
Langmuir 2004, 20, 928-933
Miltefosine-Cholesterol Interactions: A Monolayer Study I. Rey Go´mez-Serranillos,† J. Min˜ones, Jr.,† P. Dynarowicz-Ła¸ tka,*,‡ J. Min˜ones,† and E. Iribarnegaray† Department of Physical Chemistry, Faculty of Pharmacy, University of Santiago de Compostela, Santiago de Compostela, Spain, and Jagiellonian University, Faculty of Chemistry, Ingardena 3, Krako´ w 30-060, Poland Received August 5, 2003. In Final Form: October 27, 2003 Mixed Langmuir monolayers of miltefosine (hexadecylphosphocholine) and cholesterol have been investigated by recording surface pressure-area (π-A) isotherms at different subphase pHs (2, 6, and 10) and temperatures (10, 20, 25, and 30 °C). The change of both pH and temperature within the investigated range does not modify significantly the behavior of mixed films. The most pronounced effect involves condensation of the miltefosine monolayer by cholesterol, which diminishes in the following order: pH 6 > pH 2 > pH 10. The analyses of π-A and compressibility modulus dependencies indicate the existence of interactions in the investigated system; at pH 2 and 6, the strongest were found to occur for the mixed film of miltefosine molar fraction (XM) between 0.6 and 0.7 (mean value, 0.66). Such a composition corresponds to the stable complex formation wherein 2 miltefosine molecules and 1 molecule of cholesterol are strongly bound together, mainly by attractive hydrophobic forces between their apolar tails. However, at pH 10 the highest stability occurs for mixtures containing a smaller proportion of miltefosine (XM ) 0.5), which means that on alkaline subphases the ability to condense the miltefosine monolayer by cholesterol is less efficient as it requires a higher proportion of cholesterol (1:1 as compared to 1:2 at pH 2 and 6) to attain the maximum stability of the mixed film. The attractive forces between miltefosine and cholesterol are also weaker at pH 10 due to a greater solvatation of the miltefosine polar group. A similar trend is observed on increasing subphase temperature, when monolayers are more expanded.
Introduction Miltefosine (hexadecylphosphocholine, HePC), a new synthetic etherlipid derivative (Chart 1), represents the first drug to be based on a phospholipid-like structure with pronounced antineoplastic activity. It was introduced into clinical trials in the late 1980s, and topical treatment was found to be successful in breast cancer patients with skin metastases.1 So far the mode of action of miltefosine and other alkyl-lysophospholipid derivatives has not been resolved, although a number of hypotheses have been proposed,2 including disturbance of phospholipid metabolism,3 effect on membrane cholesterol,4 activation of macrophages,5 and inhibition of cellular enzymes such as protein kinase C6,7 and (Na+-K+) activated adenosine triphosphatase.6 In fact, each of these proposed mechanisms evidences a significant importance of the cellular membrane in the pharmacological action of miltefosine. The basic structure of mammalian cell membranes is a bilayer leaflet of lipids (which constitute 40% of the membrane components), stabilized by immersed proteins (50%) and carbohydrates (10%).8 In general, animal cells * Corresponding author. Tel: +48-12-6336377 ext 2236. Fax: +48-12-6340515. † University of Santiago de Compostela. ‡ Jagiellonian University. (1) Unger, C.; Eibl, H.; Breiser, A.; von Heyden, H. W.; Engel, J.; Hilgard, P.; Sindermann, H.; Peukert, M.; Nagel, G. A. Onkologie 1988, 11, 295. (2) Kosano, H.; Takatani, O. Cancer Res. 1988, 48, 6033. (3) Modolell, M.; Andreesen, R.; Pahlke, W.; Brugger, U.; Munder, P. G. Cancer Res. 1979, 39, 4681. (4) Diomede, L.; Piovani, B.; Modest, E. J.; Noseda, A.; Salmona, M. Int. J. Cancer 1991, 49, 409. (5) Yamamoto, N.; Ngwenya, B. Z. Cancer Res. 1987, 47, 2008. (6) Zheng, B.; Oishi, K.; Shoji, M.; Eibl, H.; Berdel, W. E.; Hajdu, J.; Vogler, W. R.; Kuo, J. F. Cancer Res. 1990, 50, 3025. (7) U ¨ berall, F.; Oberhuber, H.; Maly, K.; Zaknun, J.; Demuth, L.; Grunicke, H. H. Cancer Res. 1991, 51, 807. (8) Robinson, G. B. In Biological Membranes; Parson, D. S., Ed.; Plenum Press: New York, 1975.
contain three types of lipids: phospholipids, glycosphingolipids, and sterols, usually cholesterol.9 These components play a very important role in molecular recognition and intercellular communication processes. It seems fundamental for elucidating the mechanism of miltefosine action to study the interactions between miltefosine and particular membrane components, and in this connection the present work is aimed at investigating the behavior of miltefosine in mixed monolayers with cholesterol. This sterol has been chosen to study primarily for several reasons. First, it is widely distributed in all cells of the organism. Second, it is a principal component of cellular membranes and plasma lipoproteins. Specifically, in cellular membranes of liver cells,10 red cells,11,12 and myelin13 the proportion of cholesterol to phospholipids was found to be 0.83, 0.9, and 1.32, respectively. In addition, cholesterol plays an important role in modifying the physical properties of cellular membranes.14-16 Thus it seems quite appropriate to characterize the nature of miltefosine-cholesterol interaction. In this context, we have performed monolayer studies based on surface pressure-area measurements, using the Langmuir technique, upon various experimental conditions (different subphase temperatures and pHs). The interactions between miltefosine and cholesterol were examined based (9) Biological Membranes: Physical Fact and Function; Chapman, D., Ed.; Academic Press: London, 1968; p 7. (10) Dorling, P. R.; Le Page, R. N. Biochim. Biophys. Acta 1973, 318, 33. (11) Asworth, L. A. E.; Green, C. Science 1966, 151, 210. (12) Nelson, G. J. J. Lipid Res. 1967, 8, 374. (13) Demel, R. A.; London, Y.; Geuters Van Kessel, W. S. M.; Vossenberg, F. G. A.; Van Deenen, L. L. M. Biochim. Biophys. Acta 1972, 255, 321. (14) Honger, T.; Jorgensen, K.; Biltonen, R. L.; Mouritsen, O. G. Biochemistry 1996, 35, 9003. (15) Burack, W. R.; Yuan, Q.; Biltonen, R. L. Biochemistry 1993, 32, 583. (16) Yeagle, P. L. In The Biology of Cholesterol; Yeagle, P. L., Ed.; CRC Press: Boca Rato´n, FL, 1988.
10.1021/la0303254 CCC: $27.50 © 2004 American Chemical Society Published on Web 12/30/2003
Miltefosine-Cholesterol Interactions
Langmuir, Vol. 20, No. 3, 2004 929
Figure 1. Surface pressure (π)-area (A) isotherms of miltefosine, cholesterol, and their mixtures of odd and even molar fractions of miltefosine (XM) spread on water, pH 6, at 20 °C. Chart 1. Chemical Structure of Miltefosine (Hexadecylphosphocholine)
on the following dependencies: compressional modulussurface pressure, mean molecular area-composition, excess molecular area-composition, and excess free energy-composition plots. The behavior of miltefosine in mixed monolayers with other membrane lipids will be a subject of our subsequent study. Experimental Section Miltefosine (Sigma, 99%) and cholesterol (Fluka, 99%) were stored at 0 °C. The spreading solutions for Langmuir experiments were prepared by dissolving each compound in a chloroform/ absolute ethanol (Merck, p.a.) 4:1 v/v mixture (containing a few drops of amyl alcohol to facilitate spreading) with a typical concentration of ca. 0.2-0.3 mg/mL. Mixed solutions were prepared from the respective stock solutions of both compounds. The number of molecules spread on the water subphase (3.2 × 1016 molecules), with a Microman Gilson microsyringe, precise to (0.2 µL, was kept constant in all experiments. Ultrapure water (produced by a Nanopure water purification system coupled to a Milli-Q water purification system, resistivity ) 18.2 MΩ cm) was used as a subphase. The subphase temperature was controlled to within 0.1 °C by a circulating water system from Haake. To adjust the pH value of the subphase, either HCl or NaOH was added into water. The pH of aqueous subphases was measured with a Crison pH meter. Experiments were carried out with a NIMA 601 trough (Coventry, U.K.) (total area ) 525 cm2) placed on an antivibration table. Surface pressure was measured with the accuracy of (0.1 mN/m using a Wilhelmy plate made from chromatography paper (Whatman Chr1) as the pressure sensor. After spreading, the monolayers were left for 10 min for the solvent to evaporate, after which compression was initiated. Since our previous investigations indicated that miltefosine and cholesterol monolayers are nearly not influenced by such experimental conditions as surface concentration and speed of compression,17,18 we have performed our experiments for mixed monolayers using a routine compression speed of 50 cm2/min (15.6 Å2 molecule-1 min-1) and spreading always the same number of molecules at the surface (3.2 × 1016). (17) Seoane, R.; Min˜ones, J.; Conde, O.; Min˜ones, J., Jr.; Casas, M.; Iribarnegaray, E. J. Phys. Chem. B 2000, 104, 7735. (18) Rey Go´mez-Serranillos, I.; Min˜ones, J., Jr.; Dynarowicz-Ła¸ tka, P.; Iribarnegaray, E.; Casas, M. Phys. Chem. Chem. Phys., in press.
Results The surface pressure-area isotherms for both pure components, that is, cholesterol and miltefosine, have already been analyzed in detail in our former studies,17,18 and therefore herein we will concentrate on mixed systems only. For a clear representation, the isotherms recorded on water, pH 6, at 20 °C for different mole fractions of miltefosine (XM) have been separated. From Figure 1, one can clearly see that the addition of cholesterol to the miltefosine monolayer provokes its significant condensation; the limiting area of the pure miltefosine film (73.6 Å2/molecule) shifts to 44 Å2/molecule with the addition of only 0.1 mol of cholesterol to pure miltefosine. Mixed monolayers containing miltefosine within the molar fraction range of XM ) 0.1-0.7 are all shifted toward smaller areas as compared to those of both pure components. Such a behavior is characteristic of mixtures that exhibit negative deviations from ideality. If the mixture components were ideal, the mixed isotherms would be situated between those for pure components. Another characteristic feature is that the mixed monolayers of XM e 0.5 are of condensed type, similar to cholesterol, while those of XM ) 0.8-0.9 are expanded, like miltefosine. The mixtures of XM ) 0.6 and 0.7 exhibit intermediate behavior between those of condensed (solid) and expanded types and are classified as liquid-condensed (LC). The π-A isotherm of the XM ) 0.8 monolayer shows a kink at 39 mN/m and a sharp collapse at 60 mN/m. The collapse pressure of mixed films raises with increasing proportion of miltefosine in the mixed monolayer, attaining a maximum for XM ) 0.6-0.8, which accounts for their highest stability. On further increase of miltefosine concentration, the collapse pressure, πc, decreases until it reaches the value for the pure miltefosine monolayer (πc ) 39.5 mN/m). The different physical states of the monolayers and the collapse pressure (πc) values can be determined in a more precise way (as compared to the π/A isotherms) with the plot of the compressional modulus (elasticity), Cs-1, as a
930
Langmuir, Vol. 20, No. 3, 2004
Rey Go´ mez-Serranillos et al.
Table 1. Compressional Modulus (Cs-1, mN/m) and Collapse Pressure (πc, mN/m) Values for Monolayers of Cholesterol, Miltefosine, and Their Mixtures at Different Subphase pHs and Temperaturesa pH 6 10° C
20° C.
25° C
30° C
pH 2, 20° C
pH 10, 20° C
monolayer composition
Cs-1
πc
Cs-1
πc
Cs-1
πc
Cs-1
πc
Cs-1
πc
Cs-1
πc
cholesterol XM ) 0.1 XM ) 0.2 XM ) 0.3 XM ) 0.4 XM ) 0.5 XM ) 0.6 XM ) 0.7 XM ) 0.8 XM ) 0.9 miltefosine
289 248 296 329 289 257 177 76 52*-34 49 44
43.3 46.3 47.0 52.2 52.7 53.9 60.0 56.2 37.6*-59.2 38.1 36.9
280 338 320 269 260 247 129 99 53*-40 49 47
44.1 45.8 47.4 51.4 52.4 51.6 58.8 59.0 39*-59.2 39.3 39.5
280 321 325 346 255 197 120 55 48*-30 46 44
44.1 46.4 51.0 52.0 51.3 54.4 56.1 57.2 39.3*-56 37.8 37.1
338 315 270 297 273 195 145 69 45.6*-24.5 43 42
45.0 45.6 50.0 50.4 51.2 52.7 53.2 54.4 36.7*-46 36.5 35.8
379 280 342 272 273 243 168 66*-64 51*-45 50 51
43.5 44.6 49.1 51.8 51.9 51.9 57.4 37.3*-57.6 38.7*-56 38.5*-47 37.9
335 260 285 304 233 173 98 49*-52 46*-45 46 42.6
45 46.7 51.8 52.6 53.3 56.7 58.5 38*-57 37.6*-60 37.4 36.3
a
The values indicated by asterisks correspond to the first collapse.
Figure 2. Compressional modulus (Cs-1) as a function of surface pressure (π) for monolayers of miltefosine, cholesterol, and their mixtures spread on water, pH 6, at 20 °C.
function of surface pressure (π) (Figure 2). Cs-1 values were obtained by numerical calculation of the first derivative from the isotherm datapoints, according to the expression Cs-1 ) -A(dπ/dA). The highest value of the film elasticity appears as a maximum on the Cs-1 ) f(π) dependencies. Regarding the monolayer collapse, for a gradual type of collapse, the Cs-1-π curves exhibit a clear minimum. In the case when a monolayer collapse appears in the π-A isotherms as a flat plateau, πc corresponds to the surface pressure at which Cs-1 ) 0. For the cholesterol monolayer, the maximum value of Cs-1 is achieved at 10 mN/m. This maximum shifts to higher pressures with increasing proportion of miltefosine in the mixtures. For films of XM ) 0.1-0.5, the compressional modulus values at the maximum in the Cs-1 ) f(π) plots are very high (328-247 mN/m) (although they slightly diminish on increasing miltefosine proportion) and indicate their condensed (solid) state. The situation changes for films richer in miltefosine; that is, mixtures of Xm ) 0.6 and 0.7 show Cs-1 values corresponding to monolayers in a liquid-condensed state (Cs-1 values between 100 and 150 mN/m)19 while mixtures with higher miltefosine content are expanded (Cs-1 values below 100 mN/m), similar to pure miltefosine. Moreover, for mixtures (19) Davies, J. T.; Rideal, E. K. In Interfacial Phenomena, 2nd ed.; Academic Press: New York, 1963; p 265.
of XM ) 0.8 and 0.9 a characteristic sharp minimum, which occurs for both mixed films at the same surface pressure (39 mN/m), can be noticed and indicates the transition of a 2D (monolayer) state into 3D structures (film collapse). Table 1 compiles values of compressional modulus at the maximum and collapse pressures, both determined from the Cs-1-π curves for the investigated mixed systems. Apart from carrying out experiments at pH 6, we have also investigated the behavior of miltefosine-cholesterol mixtures at extreme pHs, that is, strongly acidic (pH 2) and basic (pH 10). While cholesterol is uncharged within the whole range of subphase pH, miltefosine has two ionizable (i.e., phosphate and ammonium) groups and, similar to lecithins, in the pH range 3-720-22 or 4-823,24 it should exist as a zwitterion. Our experiments carried out for both pure miltefosine monolayers18 and mixed miltefosine-cholesterol films at pH 2 and 10 did not reveal any major influence on monolayer behavior as compared to the results at pH 6. Since only slight differences are observed with changing subphase pH, instead of showing the π-A isotherms, the obtained values of Cs-1 and πc are summarized in Table 1. From the results (at 20 °C) compiled in Table 1, one may conclude the following: (i) The condensing effect of cholesterol on miltefosine monolayers (deduced by comparing the Cs-1 values) diminishes in the following sequence: pH 6 > pH 2 > pH 10. As can be seen, at pH 6, the condensation of the miltefosine monolayer includes films of XM ) 0.1-0.7 (films of XM ) 0.1-0.5 are of condensed (solid) type, and those of XM ) 0.6-0.7, liquid-condensed). At acidic pH, the condensing effect can be noticed for mixtures within the molar fraction range of 0.1-0.6 (those of XM ) 0.1-0.5 are of solid character, and the film of XM ) 0.6 is in a liquidcondensed state), while on an aqueous subphase of pH 10, the condensation occurs only for film compositions of XM ) 0.1-0.5 (0.1-0.4, solid monolayers; 0.5, liquidcondensed). (ii) The most stable monolayers (those collapsing at the highest surface pressures) are mixtures of XM ) 0.6 and 0.7 at pH 2 and 6 and of XM ) 0.5-0.6 at pH 10. (iii) The first phase transition of a monolayer (2D state) into 3D structures (first film collapse), which can be noticed as a minimum on the Cs-1-π plots, appears for mixtures of miltefosine composition XM ) 0.7 and 0.8 at pH 2 and (20) Anderson, P. J.; Pethica, B. A. In Biochemical Problems of Lipids; Popjak and Le Breton: London, 1956; p 24. (21) Gong, K.; Feng, S. S.; Lin Go, M.; Hsing Soew, P. Colloids Surf., A 2002, 207, 113. (22) Papahadjopoulos, D. Biochim. Biophys. Acta 1968, 163, 240. (23) Shah, D. O.; Schulman, J. H. J. Lipid Res. 1965, 6, 341. (24) Shah, D. O.; Schulman, J. H. J. Lipid Res. 1967, 8, 227.
Miltefosine-Cholesterol Interactions
Langmuir, Vol. 20, No. 3, 2004 931
Figure 3. Dependence of the mean molecular area (A12) on the mole fraction of miltefosine (XM) in mixed films with cholesterol, spread on an aqueous subphase of pH 2 (A), pH 6 (B), and pH 10 (C) at 20 °C.
10 but at pH 6 for XM ) 0.8 and 0.9. This transition occurs at the surface pressure of ca. 39 mN/m, which matches the collapse pressure of the pure miltefosine monolayer. To study the influence of temperature, miltefosinecholesterol mixtures were spread on water, pH 6, and the monolayers were compressed at the speed of 50 cm2/min. As can be seen in Table 1, no significant differences can be observed with the change of subphase temperature. At all the temperatures studied, monolayers of XM ) 0.1-0.5 are of solid type, that of XM ) 0.6 is always of liquidcondensed character, and mixtures containing a molar fraction of miltefosine above 0.8 are expanded. To analyze the interaction between miltefosine and cholesterol, the dependencies of mean molecular area, A12, in the mixed monolayer as a function of film composition expressed by the molar fraction of miltefosine (XM) at constant surface pressures of 5, 20, and 30 mN/m have been plotted for different subphase pHs (Figure 3). The dotted lines illustrate the additive relationship for the ideal system:25 A12 ) A1X1 + A2X2, wherein A12 is the mean molecular area per molecule in the mixed monolayer, X1 and X2 denote the mole fractions of components 1 and 2, and A1 and A2 stand for the molecular areas of the pure components at the same surface pressure at which A12 was determined. Strong negative deviations from ideal behavior can be observed for the whole mole ratio range examined, irrespective of subphase pH and temperature (results not shown here). The observed negative deviations are characteristic for contraction of the two-component film due to attractive interactions.26-28 At low surface pressures, the greatest deviations are observed over the Xm range from 0.3 to 0.6, while at higher pressures the minima are shifted toward a miltefosine composition of 0.6-0.7. The same trend can be noticed for mixed monolayers spread at lower (10 °C) and higher (25 and 30 °C) subphase temperatures (results not shown). Also, the excess area of mixing (Aexc), which can be calculated comparing the area of the mixed monolayer (25) Goodrich, F. C. In Proceedings of the 2nd International Congress on Surface Activity; Butterworth: London, 1957; Vol. 1, pp 85-91. (26) Gilardoni, A.; Margheri, E.; Gabrielli, G. Colloids Surf. 1992, 68, 235. (27) Bonosi, F.; Margheri, E.; Gabrielli, G. Colloids Surf. 1992, 65, 287. (28) Petrov, J. G.; Mobius, D.; Angelova, A. Langmuir 1992, 8, 201.
(A12) with those for the unmixed, one-component monolayers (A1 and A2), according to the equation Aexc ) A12 (A1X1 + A2X2), can provide information about the interaction between film-forming components. If the mixture is ideal or its components are immiscible, Aexc is zero and the dependence Aexc ) f(Xi) is linear. Deviations from these conditions indicate miscibility and nonideality.29,30 Figure 4 presents the results of Aexc calculated for the investigated mixtures at different subphase pHs. Negative values of Aexc prove the film condensation, or rather contraction, in the whole range of mole fractions studied. The strongest deviations occur for XM ) 0.7 at pH 2 and 6, independent of surface pressure. However, at pH 10, the most negative Aexc values appear for XM ) 0.4 in the low-pressure region and for XM ) 0.6 at higher surface pressures. In general, the most pronounced effect is observed at low π, that is, 5 mN/m, where the monolayers are in their expanded state. A more quantitative analysis of the intermolecular interactions is based on calculation of the excess free energy of mixing (∆Gexc), which can be determined from the following relation derived by Goodrich25 and Pagano and Gershfeld:31
∆Gexc ) N
∫0π(A12 - X1A1 - X2A2) dπ
(1)
Figure 5 shows ∆Gexc values at 20 °C and at surface pressures in the range of 5-30 mN/m, as a function of miltefosine concentration in the mixed monolayers. As can be seen, all values of ∆Gexc are negative, irrespective of a particular monolayer composition and surface pressure, which is consistent with thermodynamic conditions for miscibility and strong interactions in binary monolayers.25 At higher surface pressures, the minimum values of ∆Gexc appear for mixture compositions of 0.5 (pH 10) and 0.7 (pH 2 and 6). The lowest ∆Gexc value evidences the most thermodynamically stable mixture because of the strongest attractive interactions between both components. The higher the surface pressure, the stronger the interactions between mixture components. In the low (29) Costin, I. S.; Barnes, G. T. J. Colloid Interface Sci. 1975, 51, 106. (30) Bacon, K. F.; Barnes, G. T. J. Colloid Interface Sci. 1978, 67, 70. (31) Pagano, R. E.; Gershfeld, N. L. J. Phys. Chem. 1972, 76, 1238.
932
Langmuir, Vol. 20, No. 3, 2004
Rey Go´ mez-Serranillos et al.
Figure 4. Plots of the excess area of mixing (Aexc) versus mole fraction of miltefosine (XM) in mixed films with cholesterol, spread on an aqueous subphase of pH 2 (A), pH 6 (B), and pH 10 (C) at 20 °C.
Figure 5. The excess free energy of mixing (∆Gexc) as a function of mole fraction of miltefosine (XM) in mixed films with cholesterol, spread on an aqueous subphase of pH 2 (A), pH 6 (B), and pH 10 (C) at 20 °C.
Discussion
The formation of stable complexes has also been postulated for other mixed systems to explain the negative deviations from ideality.32-40 At pH 6 and 20 °C, mixed films of XM between 0.6 and 0.7 (mean value, 0.66) are most stable and exhibit the strongest contraction of the mean molecular area. Such
Mixed monolayers of miltefosine and cholesterol are more stable as compared to pure components, independent of subphase pH and temperature. This conclusion is based on the observation that mixed films collapse at higher surface pressures than films of pure components and both mean molecular areas and ∆Gexc values are negative at all compositions. It may thus be postulated that because of the existence of strong attractive interactions between film components, stable complexes between miltefosine and cholesterol are formed at the surface. These complexes are responsible for the increased stability of mixed monolayers as well as for the contraction of lift-off areas.
(32) Seoane, R.; Min˜ones, J.; Conde, O.; Casas, M.; Iribarnegaray, E. Biochim. Biophys. Acta 1988, 1375, 73. (33) Seoane, R.; Min˜ones, J.; Conde, O.; Casas, M.; Iribarnegaray, E. Langmuir 1999, 15, 3570. (34) Seoane, R.; Min˜ones, J.; Conde, O.; Iribarnegaray, E.; Casas, M. Langmuir 1999, 15, 5567. (35) Saint-Pierre-Chazalet, M.; Thomas, C.; Dupeyart, M.; Gary-Bobo, C. M. Biochim. Biophys. Acta 1988, 944, 477. (36) Gershfeld, N. L. J. Biophys. 1978, 22, 469. (37) Shah, D. O. J. Colloid Interface Sci. 1981, 79, 319. (38) Shah, D. O. J. Colloid Interface Sci. 1971, 37, 744. (39) Albrecht, O.; Gruler, H.; Sackmann, E. J. Colloid Interface Sci. 1981, 79, 319. (40) Alsina, M. A.; Mestres, C.; Valencia, G.; Garcia Anton, J. M.; Reig, F. Colloids Surf. 1989, 40, 145.
surface pressure region (5 mN/m), the minima can hardly be distinguished. No significant discrepancies can be observed at lower (10 °C) and higher (25 and 30 °C) subphase temperatures.
Miltefosine-Cholesterol Interactions
a composition corresponds to the complex stoichiometry of 2:1 (miltefosine/cholesterol). In consequence, it can be postulated that the mixed monolayer is formed by surface complexes composed of 2 miltefosine molecules and 1 molecule of cholesterol, strongly bound together, mainly by attractive hydrophobic forces between their apolar tails. In such a complex, the mean molecular area occupied by one molecule is smaller than that of either cholesterol or miltefosine. In mixed films of XM < 0.6, cholesterol occupies the same area as in a pure monolayer; however, it exerts a condensing effect on miltefosine, just as it was reported to condense monolayers from other compounds.21,32-34 The mixed film of XM ) 0.8 exhibits an anomalous surface behavior due to the presence of two collapses (Figure 1): the first one, which is seen as an inflection on the π-A isotherm, and the second one, which accounts for the final monolayer collapse. The presence of two collapses is characteristic of immiscible systems, wherein the first collapse corresponds to the squeezing out of a less stable component from the mixed monolayer, while the second corresponds to the more stable compound. In such a case, values of collapse surface pressures coincide with those for pure components. For the miltefosine-cholesterol system, the first collapse corresponds to miltefosine; however, the other one occurs at a higher surface pressure than that for pure cholesterol. The following explanation can be put forward. The mixed monolayer of XM ) 0.8 is composed of stable miltefosine-cholesterol complexes of 2:1 stoichiometry, coexisting with the excess molecules of miltefosine, which are not involved in the complex formation. Upon compression, these “free” miltefosine molecules are expelled from the mixed film at a surface pressure that is exactly the same as that for the pure miltefosine monolayer, and the remaining components are only miltefosine-cholesterol complexes which, because of their high stability, collapse at a higher surface pressure as compared to both pure components. The same process occurs for the mixed film of XM ) 0.9, in the case of which the expulsion of miltefosine is marked as a plateau at ca. 40 mN/m. The other collapse, however, is not visible because the number of miltefosine-cholesterol complexes is so small that the area corresponding to their collapse is out of the moving barrier range. In general, for each mixture containing XM > 0.7, the film components are miscible below the first collapse (surface pressures lower than 40 mN/m), and the mean molecular area is smaller than that for pure components, which evidences the presence of attractive interactions between film-forming molecules. The change of subphase pH only slightly modifies the behavior of mixed films. The most pronounced effect is observed with the condensation of the miltefosine monolayer which diminishes in the following order: pH 6 > pH 2 > pH 10. The same trend is noticed with ∆Gexc values. The observed differences, however, are not significant, and therefore it may be concluded that the interactions between film components are, in principle, of hydrophobic nature as a consequence of van der Waals attractive forces which, according to the interpretation given by other
Langmuir, Vol. 20, No. 3, 2004 933
authors for similar compounds,41,42 are established between the acyl tail of miltefosine and the steroid structure of cholesterol. However, it is possible that the polar groups of both compounds also make a contribution to the attractive interactions, thus explaining the observed differences at extreme subphase pHs. One may consider the existence of a hydrogen bond network between the polar groups of both molecules at pH 6 since, under these conditions, miltefosine is present as a zwitterion and in such a form contributes to the increased stability of the complexes. At pH 2 or 10, the polar part of miltefosine is charged (either positively or negatively), and the formation of hydrogen bonds with cholesterol OH groups is less probable. Moreover, the presence of charged polar groups in the molecule provokes their greater solvatation and, in consequence, a larger effective size. These space requirements cause the neighboring molecules to separate from each other, which weakens the cohesive interactions between their apolar tails. At acidic pH, the solvatation is smaller as compared to alkaline subphases, where the hydration occurs very easily,21 and this explains weaker interactions at pH 10 than at pH 2. The change of the effective polar group size upon pH variation, which affects the packing of the apolar tails, seems to be responsible for the particular stoichiometry of the stable miltefosine-cholesterol complexes. As mentioned above, at pH 10 the highest stability occurs for mixtures containing a smaller proportion of miltefosine (XM ) 0.5) as compared to pH 2 or 6 (XM ) 0.6-0.7). This means that on alkaline subphases the capacity to condense the miltefosine monolayer by cholesterol is less efficient as it requires a higher proportion of cholesterol (1:1 as compared to 1:2 at pH 2 and 6) to attain the maximum stability of the mixed film. The attractive forces between miltefosine and cholesterol are also weaker at pH 10 due to a greater solvatation of the miltefosine polar group. A similar trend is observed on increasing subphase temperature, when monolayers are more expanded, because thermal agitation of hydrophobic tails causes a larger separation of neighboring molecules and thus weakens the intermolecular interactions. The condensation of monolayers, qualified by the negative values of the excess molecular areas, varies with surface pressure, that is, is more pronounced at low surface pressures where films are in the expanded state and can be contracted to a higher extent than in their closely packed arrangement at higher surface pressures. The surface pressure also influences values of the excess free energy of mixing, however, in an opposite way as described above; that is, ∆Gexc values become more negative with π increase. This is quite understandable as in the condensed region, due to the smallest intermolecular distances, interactions are the strongest. Acknowledgment. This work was supported by Xunta de Galicia (Project No. PGIDT99PX12030). LA0303254 (41) Demel, R. A.; Bruckdorfer, K. R.; van Deenen, L. L. M. Biochim. Biophys. Acta 1972, 225, 311. (42) Lund-Katz, S.; Laboda, H. M.; McLean, L. R.; Phillips, M. C. Biochemistry 1988, 27, 3416.
