Effect of Dipole Potential Variations on the Surface Charge Potential of

Jan 21, 2009 - the insertion of phloretin or by the elimination of carbonyl groups at the interphase, the ... Phloretin did not cause changes in the d...
0 downloads 0 Views 121KB Size
Download PDF
J. Phys. Chem. B 2009, 113, 1607–1614

1607

Effect of Dipole Potential Variations on the Surface Charge Potential of Lipid Membranes F. Lairion and E. A. Disalvo* Laboratorio de Fisicoquı´mica de Membranas Lipı´dicas, Facultad de Farmacia y Bioquı´mica, UniVersidad de Buenos Aires, Junı´n 956 2° Piso (1113) Buenos Aires, Argentina ReceiVed: September 9, 2008; ReVised Manuscript ReceiVed: October 23, 2008

When the dipole potential of dimyristoylphosphatidylcholine (DMPC) monolayers was decreased, either by the insertion of phloretin or by the elimination of carbonyl groups at the interphase, the surface charge potential was displaced to lower negative values. At low ionic strength, the decrease of the negative charge density can be ascribed to a different exposure of the phosphate to water, as there is a good correlation to an increase in the area per lipid. At high ionic strength, the magnitude of the changes in the zeta potential produced by the effects on the dipole potential was found to be dependent on the type of anions present in the subphase. Differences between Cl- and ClO4- were ascribed to the adsorption of anions according to their different hydrations and polarizabilities. The influence of a low dipole potential on the anion adsorption can be ascribed to a less positive image charge at the membrane interior, resulting from an increase in the hydrocarbon core permittivity. This is congruent with the neutralization of interfacial dipoles and the area increase, as well as with the decrease in packing of the hydrocarbon groups. Phloretin did not cause changes in the dipole potential of dimyristoylphosphatidylethanolamine (DMPE), and in consequence, no effects on the zeta potential were measured. It is concluded that changes in the inner water/hydrocarbon plane affect the electrostatic potential measured in the outer plane of the polar headgroup region. Introduction Membrane potential relies on the membrane surface potential. The electrical field at the surface is of physiological importance in triggering or modulating catalytic processes and has an impact on channel conductivity.1,2 The surface properties of a lipid membrane comprise the surface charge and the dipole distribution. The net charge on the membrane surface generates an electric potential at the surface relative to the bulk solution (the surface or double-layer potential). This charge arises from the outer parts of the membrane such as polar head groups and the counterions adsorbing or penetrating into the Stern layer. This potential ψ(x) at plane x is given by

ψ(x) ) ψo exp(-κx)

(1)

which is referred to the outer external plane x at the membrane in contact with the fixed aqueous phase and κ-1 is the thickness of the ionic atmosphere, which is affected by the ionic strength µ ) cz2. The relation of the surface potential (ψS) to the charge density (σo) becomes

ψS ) σo /εεoκ ) σoδ/εs

(2)

where εεoκ is the diffuse double-layer capacitance (Co) and δ is the membrane interphase thickness of permittivity εs. The potential (ψS) can be reasonably equated to the zeta potential (ζ), which can be calculated from the electrophoretic mobility of liposomes in aqueous solution, using the formalisms * Address correspondence to Edgardo Anı´bal Disalvo, Laboratorio de Fisicoquı´mica de Membranas Lipı´dicas y Liposomas, Ca´tedra de Quı´mica General e Inorga´nica, Facultad de Farmacia y Bioquı´mica, Universidad de Buenos Aires, Junı´n 956 2° Piso (1113), Capital Federal, Argentina. Tel.: 0054011-4964-8249. Fax: 0054-011-4508-3653. E-mail: [email protected].

of the electrical double-layer model of Stern, Gouy, and Chapman.4-6 Alternatively, the dipole potential (ΨD) results from the orientation of constitutive dipoles, such as carbonyl and phosphate groups, normal to the water/hydrocarbon plane, the water dipoles polarized by these constitutive dipoles, and the methyl and methylene groups in the hydrocarbon core.7 Thus, ΨD is manifested between the hydrocarbon core and the first few water molecules adjacent to the lipid head groups and is positive toward the bilayer interior by several hundred millivolts.7-9 Because of this, hydrophobic anions are able to permeate the lipid bilayer.9,10 The dipole potential has been accepted to be independent of the ionic strength. However, the relation of the surface charge to the dipole potential is given in eq 2 through the surface dielectric permittivity, to which dipoles might be contributing. Moreover, the dipole potential determination in monolayers can also be affected by the presence of surface charges; for example, the interaction of ions with the head groups might imply water organization around the polar head groups that is one of the contributions to the dipole potential. This interrelation between surface charge and dipole potential is of interest from both methodological and conceptual points of view. The method used to determine dipole potential measures the potential in a monolayer as the difference in potential between an ionizing electrode above the monolayer and a reference electrode immersed in the subphase underneath it. Under such conditions, it is likely that, in addition to dipole contributions, this potential could be sensitive to changes in surface charge density produced by monolayer expansion or counterion binding. Conceptual analyses of both dipole and charge potentials have been extensively considered in the literature because of their importance in the interactions of proteins and peptides with lipid membranes.4,8,9 The total electric potential profile of a biological

10.1021/jp808007g CCC: $40.75  2009 American Chemical Society Published on Web 01/21/2009

1608 J. Phys. Chem. B, Vol. 113, No. 6, 2009 membrane surface is the result of these potential-generating phenomena. The energetics of peptide-lipid and protein-lipid interactions are dependent on the electrostatic and nonelectrostatic forces contributing to the binding with phosphatidylcholine bilayers.11,12 Alterations in the fixed charges, interfacial dipoles, and compressibility modulus can influence peptide binding.11 The effect of anions has mainly been ascribed to the Gibbs free energy of hydration and fits acceptably with the Hoffmeister effect of anions.13 Kosmotropic anions can adsorb to membranes and alter the intramembrane electric field strength.14 On the other hand, chaotropic anions such as ClO4- and SCN- induce a negative zeta potential on the surface of phosphatidylethanolamine vesicles.15 In this regard, the headgroup dipole has been predicted to act as a charge sensor, whose tilt relative to the bilayer responds as a voltmeter. This would account for changes in electrostatic potential upon ion association.16,17 Monovalent anions strongly affect conformational fluctuations of the head groups, with the magnitude of this effect following the Hoffmeister series: Cl- < Br- < NO3- < I- < SCN- < ClO4-.18 The mobility of the vesicle is directly proportional to the zeta potential, which is assumed to reflect ion-headgroup association. The effect of NaSCN and NaClO4 on the zeta potential is more than 10 times greater than that of NaCl, suggesting greater surface adsorption in the case of the chaotropic anions.15 The influence of these anions on the dipole potential has already been investigated. Changes in the dipole potential of dimyristoylphosphatidylcholine vesicles were correlated with the relative Gibbs solvation free energy of these anions, which suggested that the chaotropic (water-structure-breaking) anions (e.g., I- and ClO4-) might penetrate more deeply into the bilayer interior than do nonchaotropic anions (e.g., Cl- and Br-).13 The observations regarding the impact of large chaotropic anions on headgroup dynamics support early measurements of the electrophoretic mobility of phosphatidylcholine (PC) lipid vesicles. However, the effect of the basal dipole potential in the membrane on the adsorption of anions has not received similar attention. Specifically, the question of how the changes in the dipole potential contributions affect the charge distribution and, hence, the zeta potential has not been addressed. As inferred in eq 2, the charge potential is a function of the thickness and the dielectric constant adjacent to the surface. Thickness is directly related to the excluded volume (or hydration repulsion force) produced by water polarized at the surface, which, in turn, determines the permittivity properties. If it is thought that anions entering the Stern layer are in the region of the hydration shell of the phospholipids, the arrangements of dipole molecules can affect the permittivity and the area per lipid. In other words, properties linked to the dipole potential could affect the adsorption of anions and, hence, the charge potential. A previous work showed some correspondence between dipole and zeta potentials in lipid mixtures.19 However, simplified interface models do not consider the spatial distribution of charged sites, counterions, and co-ions with their corresponding hydration shells. In addition, it is essential to take into account that the polar head groups of the lipids are not located in a homogeneous perfect plane and that the limit between the polar and apolar region is rather diffuse. Water can penetrate to a certain extent into the glycerol region, which, in the fluid state, is manifested by a high lateral mobility of the acyl membrane components.20 Therefore, it is expected that this complexity would give rise to the cross phenomena between charge and dipole potentials inferred above.

Lairion and Disalvo As described above, the arrangement of the dipole potential makes the hydrocarbon phase positive with respect to the aqueous phase.6,8,21-23 As a consequence, large hydrophobic anions such as tetraphenylborates bind to the membrane more strongly than their positively charged analogues, such as tetraphenylphosphonium.9 Thus, the partitioning of ions could also contribute to the surface potential in relation to the relative stability of ions in solution or in the interphase region. In this regard, the terminal methyl groups of the chains might also affect the dipole potential. If, as suggested by Tatulian,24,25 the origin of the zeta potential is, in part, due to the adsorption of anions and this depends on membrane hydration, then the changes in the dipole potential should affect anion adsorption and, hence, surface potential. That is, changes in the dipole potential would modify the magnitude and possibly the charge of the Stern layer. As the dipole potential is directly related to the presence and hydration of carbonyl and phosphate groups, there are at least two possibilities to decrease the dipole potential. One is to compare lipids with and without carbonyl groups (i.e., acyl and alkyl derivatives). The other is to add molecules that can oppose their own dipoles to those of the membranes, such as phloretin.26 In addition, the contribution to the dipole potential of the headgroups can be studied by comparing dimyristoylphosphatidylcholine (DMPC) and dimyristoylphosphatidylethanolamine (DMPE), as phloretin has different influences on the dipole potential of saturated lipids according to the polar headgroup.27 The decrease in dipole potential produced by phloretin [2′,4′,6′-trihydroxy-3-(4-hydroxyphenyl propiophenone)] has been explained by the partial insertion of its own dipole at the hydrocarbon/water interface, opposing those corresponding to the lipid monolayers.28,29 Other studies have indicated that phloretin changes the transition temperature, suggesting that the action at the interface propagates into the hydrocarbon core.23 As mentioned, the surface potential of lipid membranes would be dependent on the organization of the interphase, in addition to the net electrostatic charge that the surface might bear. Thus, one or more of these effects produced by phloretin or the absence of carbonyl groups on the lipid interphase, related to a decrease in the dipole potential, could affect the adsorption of nonpermeant anions, modifying the surface charge distribution. If this is the case, then the charges adsorbed on the membrane could be affected by nonelectrostatic forces in addition to the electrostatic ones. To analyze whether the dipole potential affects the charge potential, this article reports the effects of phloretin and ether derivatives on the adsorption of anions of different polarizabilities and, hence, on the zeta potential. Studies of the area per lipid, by determining the surface pressure in monolayers and phase transition profile by measuring the fluorescence anisotropy of dimyristoylphosphatidylcholine (DMPC) and dimyristoylphosphatidylethanolamine (DMPE), were carried out to determine the influence of the dipole potential on the surface charge of liposomes with the same composition and different ionic strengths of Cl- and ClO4-. In addition, as the dipole potential is directly related to the presence of carbonyl groups in the ester union, a comparison with dialkyl phosphatidylcholines (PCs) and dialkyl phosphatidylethanolamines (PEs) was done. With this purpose, a model defining the interphase region between an inner plane corresponding to the hydrocarbon/water interface and an outer plane defined by the shear plane of the zeta potential was used.

Surface Charge Potential of Lipid Membranes

J. Phys. Chem. B, Vol. 113, No. 6, 2009 1609

Materials and Methods Lipids and Drugs. Dimyristoylphosphatidylcholine (DMPC), 1,2-di-O-tetradecyl-sn-glycero-3-phosphocholine (etherPC), dimyristoylphosphatidylethanolamine (DMPE), and 1,2-di-Otetradecyl-sn-glycero-3-phosphoethanolamine (etherPE) were obtained from Avanti Polar Lipids, Inc. (Alabaster, AL) and used as received. The purity of the lipids was checked by thinlayer chromatography using a chloroform/methanol/water mixture as running solvent. Phloretin (Phl) and the fluorescence probe 1,6-diphenyl-1,3,5hexatriene (DPH) were obtained from Sigma-Aldrich (Saint Louis, MO) and used as received. Chloroform, KClO4, and KCl were analytical grade. Water was of MilliQ quality. Liposome Preparation. Lipids, pure or in mixtures with phloretin, dissolved in a chloroform solution, were dried under N2 stream. Multilamellar liposomes (MLVs) were prepared by dispersing the dry lipid films in water (fluorescence experiments) or buffer Hepes 1 mM + KCl or KClO4 in different ratios (for zeta potential experiments) at temperatures above the phase transition, for 60 min. In all cases the relation lipid to water was maintained in order to make the preparation more reproducible. Zeta Potential. MLVs prepared as described were used to determine the zeta potential (ζ) in a Zeta-Meter System 3.0 instrument at 20 ( 2 °C. The voltage was fixed at 75 V. The total lipid concentration in all cases was 52 µM. The phloretin/lipid ratios tested were 0.075:1 and 0.15:1. Reported data are the averages of 20 measurements performed for each set of conditions assayed, with at least five different batches of liposomes. The zeta potential (ζ) is expressed as a function of the viscosity (η), the dielectric constant of the suspension (ε), and the electrophoretic mobility (u) as

ζ ) ηu/ε

(3)

for particles in which the ionic atmosphere is relatively thin compared with the radius of the particle (a) (κa . 1). The electrophoretic mobility (u) for spherical particles is

u ) V/E ) Q/6πηa

(4)

where V is the electrophoretic velocity in terms of the electrical force on the particle (E) and the particles’ charge (Q), η is the viscosity, and a is the radius of the particle. Fluorescence Anisotropy. MLVs were prepared as previously described, with the addition of the fluorescence probe DPH to the chloroform lipid solution, in a probe/lipid ratio of 1:300. Preparations were protected from light. Fluorescence anisotropy (r) was determined with a PerkinElmer LS55 Luminiscence spectrometer. The excitation and emission wavelengths (λ) were 350 and 452 nm, respectively. The temperature was controlled by an external system to within (0.2 °C. The total lipid concentration in all cases was 0.1 mg per milliliter of liposome suspension, and the phloretin/lipid ratio was 0.075:1. Reported data are the averages of 10 measurements for each set of conditions and temperature, with at least three different batches of liposomes. Calculation of Area per Lipid by Measures of the Surface Pressure in Monolayers. The formation of lipid monolayers with and without phloretin was monitored by measurements of the surface pressure of the different lipid monolayers in a Kibron µtrough S instrument at constant temperature and area.

TABLE 1: Effect of the Variations in the Dipole Potential on the Zeta Potential of PCs and PEs at 20 °C

DMPC DMPC + phloretin (0.15 molar ratio) di(etherPC) DMPE DMPE + phloretin (0.15 molar ratio) a

dipole potential (SDa (mV)

zeta potential (SDa (mV)

515.5 ( 21.0 380.5 ( 12.3

-13.7 ( 1.2 -9.7 ( 1.0

437.0 ( 23.6 568.6 ( 13.7 547.4 ( 18.9

+1.8 ( 1.2 -45.3 ( 0.9 -43.5 ( 2.2

SD ) standard deviation.

Aliquots of a chloroform solution of pure lipids or lipids in mixtures with phloretin were spread on a clean surface of water and left to reach constant surface pressures, until no changes were observed with further additions of lipid (saturation). The resulting surface pressures are expressed in units of millinewtons per meter (mN/m). Under these conditions, the measurements are obtained with lipids in the monolayer in equilibrium with liposomes in the subphase. The lipid conformations stabilized spontaneously according to the aqueous solution properties, without forcing by the application of a lateral pressure. The saturation points of the different monolayers, under the conditions assayed, were determined by considering the standard deviation of the results at the plateau of the curve (see Figure 5 below). Those points for which the difference from the mean point of saturation was higher than the standard deviation were not considered as the first point of saturation. With these criteria, areas per lipid were calculated with the first point of the saturation plateau of a curve of monolayer surface pressure vs number of nanomoles of lipid added to a constant area of the trough. Considering that each aliquot corresponds to 1 nmol, each determination was affected by an error corresponding to (0.5 nmol. The error in the area is reported in Table 2 below. Dipole Potential in Monolayers. Dipole potential (ΨD) was determined when the surface pressure of the monolayers reached the saturation point described above. The values of the interfacial potential were determined through a circuit of high impedance, connecting an ionizing electrode above the monolayer and a reference electrode in the aqueous subphase, using the expression

Vsurf)VAg/AgCl - Vgrd

(5)

where Vsurf is the potential of the clean aqueous surface, measured as the potential difference between a Ag/AgCl reference electrode immersed in the solution underneath the surface (VAg/AgCl) and a grid displaced ∼2 mm above the surface (Vgrd). This grid was the sensor of an ionizing electrode that emits alpha particles in order to achieve an electrical connection across the air. The dipole potential of the monolayer (ΨD) was evaluated as

ψD)Vlip - Vsurf

(6)

where Vsurf is the potential of the clean surface (without lipids) described above and Vlip is the potential measured with the same setup after the lipid monolayer had formed on the air-water interphase. For these measurements, aliquots of a chloroform solution of pure lipids or mixtures of lipids and phloretin at

1610 J. Phys. Chem. B, Vol. 113, No. 6, 2009

Lairion and Disalvo TABLE 2: Effect of Phloretin on the Surface Pressure (mN/m) and Area Per Molecule (Å2/molecule) of DMPC, DMPE, and Their Ether Derivativesa water

Phl (1:1) 2

nmol of Π Å2/ nmol of Π Å/ lipids saturation molecule lipids saturation molecule DMPC etherPC DMPE etherPE a

5 4 6 6

48 48 45 44.5

67.5 84.7 56.1 56.3

((6.7) ((10.6) ((4.7) ((4.7)

4 3 6 6

49 49 44 45.5

84.4 ((10.6) 112.9 ((18.8) 56.1 ((4.7) 56.3 ((4.7)

All measurements conducted at 28 °C.

Figure 1. Effects of the phloretin on the dipole and zeta potentials of DMPC membranes: (9) zeta potential (ζ), (∆) dipole potential (ΨD) at saturation pressures, (0) dipole potential (ΨD) at a surface pressure of 20 mN/m.

different ratios were added to the interface of the aqueous phase until a constant surface pressure was reached. Temperature was set at the values indicated and measured with a calibrated thermocouple immersed in the subphase and maintained to within (0.5 °C. A solution of 1 mM KCL was used in the subphase in all cases. Therefore, the reference potential is constant at the temperature of measurement. Results As pointed out in the Introduction, one contribution to the dipole potential of a lipid membrane is associated with the water/ lipid interphase and is attributed to the CdO and PdO groups and the water organized by them. The contribution of the carbonyls can be eliminated in the alkyl derivatives or counteracted by the addition of phloretin to DMPC membranes. When inserted into the monolayer, phloretin decreases the dipole potential of DMPC at a surface pressure of ca. 48 mN/m (which corresponds to a saturated monolayer). To compare the effect of phloretin between monolayers and bilayers, the changes were followed in a monolayer at a surface pressure between 20 and 30 mN/m (a value comparable to that reported for bilayer liposomes). In Figure 1, it is observed that the decrease in the dipole potential corresponds to a decrease in the negative surface charge potential at similar surface pressures. In this case, the zeta potential of DMPC liposomes at 20 °C was determined in 1 mM KCl, the same solution on which the monolayers were formed. A decrease in the dipole potential was observed with an increase in the molar ratio of phloretin, in parallel with the shift of the zeta potential to less negative values, reaching a saturation value at around -8 mV, at a 0.3 molar ratio. The less negative values of the zeta potential are certainly a consequence of the decrease in the negative surface charge at the liposome surface. As observed in Table 1, the decreases in the dipole potential in PCs of 78.5 and 135 mV were correlated with positive shifts of 15 and 4 mV in the zeta potential. However, in PEs, the variation in the dipole potential was within the standard deviation, and no significant change in the zeta potential was measured. This response was observed for both Cl- and ClO4anions. This is congruent with previous results showing that phloretin has no effect on the dipole potential of DMPE. This

Figure 2. Changes in the zeta potential (ζ), as a function of (A) KCl and (B) KClO4 concentrations: (9, solid line) pure DMPC, (0, line) pure etherPC. In all cases, the standard deviation was lower than 10%.

invariance is another indication of the correlation between dipole potential and zeta potential for lipid membranes. The shift of the surface potential of DMPC liposomes to negative values is attenuated by an increase in KCl concentration (Figure 2A). However, concentration has no effect on the surface potential of the alkylPC, which remains around 0 mV, within the experimental error. A noticeably different effect was found when the ionic strength was increased with KClO4. In this case, the zeta potential of both the esterPC and etherPC liposomes displaces to more negative values (Figure 2B). The increase in ionic strength with KCl for liposomes with phloretin/DMPC molar ratios of 0:1, 0.075:1, and 0.15:1 shifts the zeta potential in the same direction and reaches the same saturation value at -6/-7 mV. At ionic strengths below 3 mM, phloretin decreases the zeta potential of DMPC liposomes. However, no change was observed at high ionic strength of KCl (Figure 3A).The ionic strength did not significantly affect the zeta potential throughout the concentration range when phloretin was added to etherPC (Figure 3B). Figure 4A shows that zeta potential changed toward negative values in pure DMPC liposomes when the ionic strength was increased with KClO4, but the presence of phloretin attenuated this shift. Figure 4B shows a similar trend for etherPC. The values of the area per lipid for DMPC and DMPE and their alkyl derivatives with and without phloretin, obtained at saturation of the lipid monolayer, are summarized in Table 2. They were obtained from plots (data not shown) similar to those in Figure 5, which shows the surface pressures at different concentrations of DMPC or DMPC/Phl mixtures added to the trough. To visualize the effect on the dipole potential of the carbonyl deletion in the ether-linked PC, independently of the area change, we also measured the dipole potential of ether

Surface Charge Potential of Lipid Membranes

Figure 3. Changes in the zeta potentials (ζ) of (A) DMPC and (B) etherPC with and without phloretin, as a function of KCl concentration. (A) (9, solid line) Phl/DMPC ) 0:1, (9, dotted line) Phl/DMPC ) 0.075:1, (0, solid line) Phl/DMPC ) 0.15:1. (B) (b, solid line) Phl/ etherPC ) 0:1, (b, dotted line) Phl/etherPC ) 0.075:1, (O, solid line) Phl/etherPC ) 0.15:1. In all cases, the standard deviation was lower than 10%.

Figure 4. Changes in the zeta potentials of (A) DMPC and (B) etherPC with and without phloretin as a function of KClO4 concentration. (A) (9, solid line) Phl/DMPC ) 0:1, (9, dotted line) Phl/DMPC ) 0.075: 1, (0, solid line) Phl/DMPC ) 0.15:1. (B) (9, solid line) Phl/etherPC ) 0:1, (9, dotted line) Phl/etherPC ) 0.075:1, (0, solid line) Phl/ etherPC ) 0.15:1. In all cases, the standard deviation was lower than 10%.

J. Phys. Chem. B, Vol. 113, No. 6, 2009 1611

Figure 6. Effect of phloretin on the phase properties of DMPC, as measured by DPH anisotropy: (9) DMPC, (0) DMPC/Phl ) 1:0.075. In all cases, the standard deviation was lower than 5% of the measured data.

Figure 7. Effect of phloretin on the phase properties of DMPE, as measured by DPH anisotropy: (b) DMPE, (O) DMPE/Phl ) 1:0.075. In all cases, the standard deviation was lower than 5% of the measured data.

at 28 °C. This indicates that the absence of CO groups decreases the dipole potential independently of the area increase. In Figure 5, the amount of lipids required to saturate a monolayer of pure DMPC on the surface of an aqueous solution, at 28 °C, decreased upon inclusion of phloretin. This implies that phloretin increases the area per molecule of this phospholipid. It can be observed that the area per lipid for DMPC/ phloretin was higher than that for pure DMPC and comparable to that obtained for pure etherPC (Table 2). No effect on the area due to phloretin addition was observed for DMPE and etherPE.

Figure 5. Effect of phloretin on the surface pressure and saturation values of lipid monolayers of DMPC spread at an air-water interphase: (9) DMPC, (0) DMPC and phloretin

and ester derivatives at similar chosen areas. The results were dipole potentials of 435.2 ( 38.1 mV at 168.8 Å2/DMPC molecule and 356.7 ( 28.4 mV at 169.3 Å2/etherPC molecule,

On the other hand, the inclusion of phloretin in liposomes of DMPC (0.075:1 molar ratio) did not significantly affect the transition temperature of DMPC, as seen with fluorescence anisotropy of DPH, but decreased the cooperativity of the transition (Figure 6). No effect was seen on DMPE liposomes, where there is no interaction between phloretin and the phospholipids (Figure 7). However, the phase of both types of membranes appeared to be the same at the temperature at which the zeta potential was measured, ca. 20 °C.

1612 J. Phys. Chem. B, Vol. 113, No. 6, 2009 Discussion As was mentioned previously, the interphase of a lipid membrane can be defined as a region between two planes. An inner plane at the glycerol backbone is defined by the water/ hydrocarbon interface, which allows the orientation of the CO groups to be defined as normal or parallel to this plane. This is identified as the Gibbs dividing plane, which refers to the ideal boundary between the hydrocarbon and the polar region. The outer plane corresponds to the ideal plane tangential to the polar head groups to which the zeta potential measurements are referred. The distribution of dipoles with respect to the inner plane determines the dipole potential. In turn, the plane dividing the polar region from the aqueous bulk phase can be identified as the plane at which the mobile phase (the liposome) and the immobile phase (the aqueous solution) coincide. The potential at the slipping plane corresponds to the zeta potential. Between the two planes, a bidimensional solution of hydrated polar groups can be located.30 This region is ∼5.1-5.3 Å in thickness and amounts 14-18 water molecules per lipid.31,32 It has been argued that lipids in a monolayer are in equilibrium with lipids in the liposomes formed in the subphase. Therefore, the chemical potentials of the lipids are equal, and the comparison is straightforward.20 With this argument, the comparison of the dipole potentials of the lipid monolayer and zeta potential of liposomes (at similar surface pressures) is valid; see Figure 1. The results of the present work indicate that the absence of carbonyl groups or the insertion of phloretin in the lipid membrane, each of which promotes a decrease in the dipole potential, changes the potential at the outer membrane plane. In other words, it suggests that the dipole and the zeta potentials might be interdependent (Figure 1 and Table 1). As shown by the relative increase of the phloretin molar ratio or the absence of carbonyls, the decrease in the dipole potential is linked to a decrease in the negative surface charge at the liposome surface. In the case in which no changes in the dipole potential are observed, as in the case of the addition of phloretin to DMPE (condensed phase), no effect on the zeta potential is apparent. Without phloretin, the zeta potential of neutral lipids such as dimyristoylphosphatidylcholines (Figures 1 and 2A), determined by means of the electrophoretic mobility of liposomes (see eqs 2 and 3), is negative at moderate ionic strengths. For etherPC, the zeta potential remains unchanged with the ionic strength, near 0 mV. EtherPC has a lower dipole potential than DMPC, because of the absence of carbonyls (among other reasons), and this is coincident with a less negative surface charge potential. The origin of the negative charge in PC membranes has been a matter of debate. On one hand, it has been ascribed to the exposure of the negative phosphate groups to the aqueous media in the outer plane, with respect to that containing the positively charged choline groups.33 Another interpretation is that the negative charges result from the adsorption of negative anions of the salt solution in which the lipids were dispersed.24,25 The present results showing parallel changes in the dipole and surface charge potentials can be interpreted in light of the two proposals at different ionic strengths. At low ionic strength (1 mM), the decrease of the dipole potential by the elimination of the carbonyl group produces a shift of the zeta potential to less negative values for both Cland ClO4- (Figure 2A,B). As the origin of the negative charges at the membrane interphase can be ascribed to two possibilities, the correlation

Lairion and Disalvo of the dipole potential decrease with the change in zeta potential should be compared with them both. If the negative charges are due to the exposure of the phosphate group to the aqueous medium, it is known that the phosphates in the etherPC are more hydrated than those corresponding to esterPC.34-36 Therefore, the negative charge of the phosphate would be more screened by the dielectric permittivity of the water solution around the phosphate, decreasing the net surface charge density. On the other hand, if the zeta potential is due to the adsorption of anions, the decrease in the dipole potential would decrease the positive image charge potential inside the membrane, thus reducing their adsorption. At high ionic strength, a different picture is found for the two lipids depending on the anion tested. In the presence of Cl-, the ionic strength changes the zeta potential of DMPC to positive values, whereas for etherPC, the zeta potential remains unchanged (Figure 2A). In contrast, a similar effect is observed for the two lipids in the presence of ClO4- (Figure 2B). Under these conditions, the shift of zeta potential to negative values indicates a significant adsorption, independent of the lipid, due to the high polarizability of the anion. The decrease to less negative values for DMPC, promoted by the increase in ionic strength with KCl, suggests a screening of the negative charges by K+ that cannot be counteracted by Cl- adsorption because of its low polarizability. Instead, no effect is observed for etherPC because of the absence of net charges, as deduced by the zeta potential value near zero. A further observation of the data in Figure 2A,B indicates that the shift to negative values from etherPC to DMPC at low ionic strength (1 mM Cl-) could depend on the fixed charges. For DMPC, the lower exposure of the phosphate to the aqueous phase reduces the effect of water permittivity, thus increasing the net negative charge. At the same point, the presence of ClO4reduces the zeta potential from -13 to -16 mV (Figure 2A,B at 1 mM), indicating the contribution of anion adsorption. In addition, in the case of etherPC (Figure 2A), the greater decrease when ClO4- is added (-10 mV, Figure 2B) would indicate that anion adsorption is favored in the absence of the repulsive effect of the net charges. When the dipole potential is reduced by the incorporation of phloretin into DMPC membranes, the surface potential becomes less negative (Figures 1 and 3A). That is, in relation to the two interpretations given above, phloretin either hinders the charge of the phosphate groups or decreases the adsorption of anions. When the exposure of the phosphate is high, as a consequence of the high dielectric constant of water, the net charge is reduced concomitantly. That is, less exposure to water increases the negative charge density, and high exposure reduces the negative charge density. However, the increase of the ionic strength with KCl attenuates the zeta potential of DMPC membranes to less negative values, with and without phloretin (Figure 3A). This implies that negative charges are screened under both conditions. The inclusion of phloretin not only reduces the dipole potential but also increases the area per lipid (Table 2 and conclusions below), producing an effect similar to that produced by the absence of CO groups. Thus, the increase in area increases the screening of the negative charges, and the addition of more salt would produce a much lower (to less negative) value in the zeta potential, up to 3 mM. As the same saturation value is reached for all phloretin/lipid ratios (over 5 mM), it can be thought that the decrease in the zeta potential, by the screening of the phosphates by the cations, is compensated by the adsorption of Cl- anions (which counteracts the decrease in the surface density of the fixed

Surface Charge Potential of Lipid Membranes

J. Phys. Chem. B, Vol. 113, No. 6, 2009 1613

charges). This renders the surface potential at high KCl concentration unchanged with and without phloretin. No significant changes are observed for the zeta potential of etherPC, with or without phloretin, which remains around 0 mV (Figure 3B). That is, even though phloretin also causes an area increase, this does not imply a further exposure of the phosphates to water beyond that produced in the etherPC. The possibility of anion adsorption appears reaffirmed by the effect found with highly polarizable anions, such as ClO4(Figure 2B). In this case, liposomes made of pure lipids become more negative in the presence of ClO4-, than in the presence of Cl- (indicating a significant adsorption, independent of the lipid, as a result of the high polarizability of the anion). When phloretin is added, both become less negative (that is, in correspondence with the two interpretations, phloretin either hinders the charge of the phosphate groups or decreases the adsorption of anions). Nevertheless, with an increase in the ionic strength of ClO4-, no difference is observed in the two lipids, showing prevalence for the adsorption of anions. The decrease of anion adsorption in the presence of phloretin could be ascribed, as was mentioned, to changes in area and anisotropy produced by its insertion. In previous works, we have shown that phloretin decreases the dipole potential of DMPC and di (ether) PC, but does not affect the homologues of phosphatidylethanolamines.37,38 This is indicative of a direct interaction of the molecule with the phosphate groups in phosphatidylcholines, as a result of its higher exposure to the water phase. The pronounced downward shift on the asymmetric vibration frequencies of the phosphates, shown by FTIR spectroscopy, correlates with dehydration of these groups.37-39 The expansion in the area observed in this work for DMPC and etherPC, but not for DMPE and etherPE, is congruent with those results. This increase in the area in the monolayers is in correlation with the observed decrease in the dipole potential and congruent with the interaction with the phosphate groups exposed to the water phase in bilayers. However, the strong lipid-lipid interaction between the PE polar head groups hinders phloretin penetration. Parallel to the area changes, the presence of phloretin leads to changes in the membrane organization at the inner plane, as shown by anisotropy determinations with DPH, which senses the ordering at the hydrocarbon core. This is another change affecting the dipole potential contribution. We observed a change in the phase transition profile of DMPC liposomes, with a decrease in the cooperativity. In other words, the effect of phloretin is not restricted to the interphase. Area expansion and the effect of hydrocarbon phase were not observed with the DMPE liposomes. The changes in the hydrocarbon phase are congruent with the changes in the dipole potential, as this potential is positive with respect to the outer aqueous phase. The higher the charge (q), the greater the dipole potential, and the more extensive the adsorption and permeation of negatively charged ions, depending on their polarizability.6,9,10 According to the Born energy (W), the transfer of an ion of radius R into the hydrocarbon phase from water is given by the difference in the permittivities of the adjacent water medium (εaq) and the hydrocarbon phase (εh)

W ) (q2 /8πεo)(1/R)(1/εh - 1/εaq)

(7)

This energy accounts for the permeation of anions with hydrophobic groups and decreases with phloretin, because the

inner phase is less positive. Therefore, anions would be less favored to be transferred. In the case of simple anions, it is possible that they could be adsorbed into the interphase because they cannot penetrate the hydrocarbon core.12 The reduction of the hydrocarbon chain packing, as denoted by the less steep change in anisotropy, is congruent with the area increase. Thus, the dielectric constant εh ) 2 of the hydrocarbon core is increased by water entrance (eq 7), decreasing the adsorption of polarizable anions. The fact that phloretin decreases the order suggests that permittivity increases and adsorption decreases. This could be explained by a higher water penetration into the hydrocarbon core when the membrane expands due to phloretin.20 This would affect the thickness (δ) and the permittivity (εs) in eq 2. Conclusions The decrease in dipole potential decreases the negative charges at the membrane surface, which is reflected in a less negative zeta potential. The negative potentials are due to fixed charges, such as the phosphate group at the interphase region, and to adsorbed counterions. The first are more relevant at low ionic strength, and the second predominate at ionic strengths higher than 3 mM. The removal of carbonyl groups, when comparing esterPC and etherPC at constant area, or the inclusion of phloretin in DMPC membranes decreases the negative charge density at low ionic strength. At high ionic strength, the anion adsorption is attenuated to different extents depending on the polarizability of the anion. In all cases, the influence of the decrease in dipole potential on anion adsorption can be ascribed to a less positive image charge at the membrane interior due to an increase in the hydrocarbon core permittivity. This is congruent with the decrease in packing observed by DPH. Acknowledgment. This work was supported with funds from Agencia Nacional de Promocio´n Cientı´fica y Te´cnica (PICT 0324R), CONICET (PIP 5476), and UBACyT (B047). E.A.D. is member of the research career of CONICET (R. Argentina). References and Notes (1) Kinnunen, P. K. J. Lipid bilayers as osmotic response elements. Cell. Physiol. Biochem. 2000, 10, 243–250. (2) Aguilella, V. M.; Bezrukov, S. M. Alamethicin channel conductance modified by lipid charge. Eur. Biophys. J. 2001, 30, 233–241. (3) Kleijn, J. M.; van Leeuwen, H. P. Electrostatic and electrodynamic properties of biological membranes. In Physical Chemistry of Biological Interfaces; Baszkin, A., Norde, W., Eds.; Marcel Dekker: New York, 2000; pp 49-84. (4) McLaughlin, S. The electrostatic properties of membranes. Annu. ReV. Biophys. Biophys. Chem. 1989, 18, 113–136. (5) Murray, D.; Arbuzova, A.; Honig, B.; McLaughlin, S. The role of elecrostatic and nonpolar interactions in the association of peripheral proteins with membranes. In Peptide-Lipid Interactions; Simon, S. A., McIntosh, T. J., Eds.; Current Topics in Membranes; Academic Press: New York, 2002; Vol. 52, Chapter 10, pp 277-307. (6) Haydon, D. A.; Hladky, S. B. Ion transport across thin lipid membranes: A critical discussion of mechanisms in selected systems. Q. ReV. Biophys. 1972, 5, 187–282. (7) Smaby, J. M.; Brockman, H. L. Surface dipole moments of lipids at the argon-water interface: Similarities among glycerol-ester-based lipids. Biophys. J. 1990, 58, 195–204. (8) Brockman, H. L. Dipole potentials of lipid membranes. Chem. Phys. Lipids 1994, 73, 57–79. (9) Cafiso, D. Influences of charges and dipoles on macromolecular adsorption and permeability. In Permeability and Stability of Lipid Bilayers; Disalvo, E. A., Simon, S. A., Eds.; CRC Press: Boca Raton, FL, 1995; Chapter 9, pp179-195.

1614 J. Phys. Chem. B, Vol. 113, No. 6, 2009 (10) Flewelling, R. F.; Hubbell, W. L. The membrane dipole potential in a total membrane potential model: Applications to hydrophobic ion interactions with membranes. Biophys. J. 1986, 49, 541–552. (11) McIntosh, T. J.; Vidal, A.; Simon, S. A. The energetic of peptidelipid interactions: Modulation by interfacial dipoles and cholesterol. In Peptide-Lipid Interactions; Simon, S. A., McIntosh, T. J., Eds.; Current Topics in Membranes; Academic Press: New York, 2002; Vol. 52, Chapter 11, 309-338. (12) Martini, M. F.; Disalvo, E. A. Effect of polar head groups on the activity of aspartyl protease adsorbed to lipid membranes. Chem. Phys. Lipids 2003, 122 (1-2), 177–183. (13) Clarke, R. J.; Lupfert, C. Influence of anions and cations on the dipole potential of phosphatidylcholine vesicles: A basis for the Hofmeister effect. Biophys. J. 1999, 76, 2614–2624. (14) Koynova, R.; Brankov, J.; Tenchov, B. Modulation of lipid phase behavior by kosmotropic and chaotropic solutes: Experiment and thermodynamic theory. Eur. Biophys J 1997, 25, 261–274. (15) McLaughlin, S. Electrostatic potentials at membrane-solution interfaces. Curr. Top. Membr. Transp. 1977, 9, 71. (16) Sachs, J. N.; Hirsh, N.; Petrache, H. I.; Woolf, T. B. Changes in Phosphatidylcholine Headgroup Tilt and Water Order Induced by Monovalent Salts. Molecular Dynamics Simulations. Biophys. J. 2004, 86, 3772– 3782. (17) Akutsu, H.; Nagamori, T. Conformational analysis of the polar head group in phosphatidylcholine bilayers: A structural change induced by cations. Biochemistry 1991, 30, 4510–4516. (18) Rydall, J. R.; Macdonald, P. M. Investigation of anion binding to neutral lipid membranes using 2H NMR. Biochemistry 1992, 31, 1092– 1099. (19) Luzardo, M. C.; Peltzer, G.; Disalvo, E. A. Surface potential of lipid interfaces formed by mixtures of phosphatidylcholine of different chain lengths. Langmuir 1998, 14 (20), 5858–5861. (20) Simon, S. A.; McIntosh, T. J. The depth of water penetration in lipid bilayers. In Methods in Enzymology; Packer, L., Deamer, D., Eds.; Academic Press: New York, 1985; pp 511-521. (21) Clarke, R. J. The dipole potential of phospholipid membranes and methods for its detection. AdV. Colloid Interface Sci. 2001, 89-90, 263– 281. (22) Cseh, R.; Benz, R. The adsorption of phloretin to lipid monolayers and bilayers cannot be explained by Langmuir adsorption isotherms alone. Biophys. J. 1998, 74, 1399–1408. (23) Cseh, R. R.; Benz, R. Interaction of phloretin with lipid monolayers: Relationship between structural changes and dipole potential change. Biophys. J. 1999, 77, 1477–1488. (24) Tatulian, S. Effect of lipid phase transition on the binding of anions to dimyristoylphosphatidylcholine liposomes. Biochim. Biophys. Acta 1983, 736 (2):), 189–195.

Lairion and Disalvo (25) Tatulian, S. Binding of alkaline-earth metal cations and some anionsto phosphatidylcholine liposomes. Eur. J. Biochem. 1987, 170 (12), 413–420. (26) Simon, S. A.; McIntosh, T. J.; Magid, A. D.; Needham, D. Modulation of the interbilayer hydration pressure by the addition of dipoles at the hydrocarbon/water interface. Biophys. J. 1992, 61, 786–799. (27) Lairion, F.; Disalvo, E. A. Effect of phloretin on the potential of phosphatidylcholine, phosphatidylethanolamine and phosphatidylglycerol monolayers. Langmuir 2004, 20, 9151–9155. (28) Bechinger, B.; Seelig, J. Interaction of electric dipoles with phospholipid head groups. A 2H and 31P NMR study of phloretin analogues in pohsphatidylcholine membranes. Biochemistry 1991, 30, 3923–3929. (29) Andersen, O.; Finkelstein, A.; Katz, I.; Cass, A. Effect of phloretin on the permeability of thin lipid membranes. J. Gen. Physiol. 1976, 67, 749–771. (30) Defay, R.; Prigogine, I. Surface Tension and Adsorption; J. Wiley and Sons: New York, 1966. (31) Simonn, S. A.; McIntosh, T. J. Magnitude of the solvation pressure depends on dipole potential. Proc. Natl. Acad. Sci. 1989, 86 (23), 9263– 9267. (32) Disalvo, E. A.; de Gier, J. Contribution of aqueous interphases to the permeability barrier of lipid bilayers. Chem. Phys. Lipids 1983, 32, 39– 47. (33) Makino, K.; Yamada, T.; Kimura, M.; Oka, T.; Ohshima, H.; Kondo, T. Temperature and ionic strength-induced conformational changes in the lipid head group region of liposomes as suggested by zeta potential data. Biophys. Chem. 1991, 41, 175–183. (34) Lewis, R.; Pohle, W.; McElhaney, R. The interfacial structure of phospholipids bilayers: Differential scanning calorimetry and Fourier transform infrared spectroscopic studies of 1,2-dipalmitoyl-sn-glycerol-3phosphorylcholine and its dialkyl and acyl-alkyl analogs. Biophys. J. 1996, 70, 2736–2746. (35) Lairion, F.; Disalvo, E. A. Effect of arbutin on the dipole potential and area per lipid of ester and ether phosphatidylcholine and phosphatidyl ethanolamine monolayers. Biochim. Biophys. Acta 2007, 1768, 450–456. (36) Lairion, F.; Disalvo, E. A. Effect of trehalose on the contributions to the dipole potential of lipid monolayers. Chem. Phys. Lipids 2007, 150, 117–124. (37) Disalvo, E. A.; Lairion, F.; Diaz, S.; Arroyo, J. Physical chemistry of lipid interfaces: State of hydration, topological and electrical properties. In Recent Research DeVelopments in Biophysical Chemistry; Condat, C. A., Baruzzi, A., Eds.; Research Signpost: Kerela, India, 2002; pp 181-197. (38) Disalvo, E. A.; Lairion, F.; Martini, F.; Almaleck, H.; Diaz, S.; Gordillo, G. Water in biological membranes at interfaces: Does it play a functional role. J. Argentine Chem. Soc. 2004, 92 (4/6), 1–22. (39) Diaz, S.; Lairion, F.; Arroyo, J.; Biondi de Lopez, A. C.; Disalvo, E. A. Contribution of phosphate groups to the dipole potential of dimyristoylphosphatidylcholine membranes. Langmuir 2001, 17 (3), 852–855.

JP808007G