Adsorption of divalent cations to bilayer membranes containing phosphatidylserine

The Stern equation, a combination of the Langmuir adsorption isotherm, the Boltzmann relation, and the Grahame equation from the theory of the diffuse double layer, provides a simple theoretical framework for describing the adsorption of charged molecules to surfaces. The ability of this equation to describe the adsorption of divalent cations to membranes containing brain phosphatidylserine (PS) was tested in the following manner. Charge reversal measurements were first made to determine the intrinsic 1:1 association constants of the divalent cations with the anionic PS molecules: when the net charge of a PS vesicle is zero one-half of the available sites are occupied by divalent cations. The intrinsic association constant, therefore, is equal to the reciprocal of the divalent cation concentration at which the mobility of a PS vesicle reverses sign. The Stern equation with this association constant is capable of accurately describing both the zeta potential data obtained with PS vesicles at other concentrations of the divalent cations and the data obtained with with vesicles formed from mixtures of PS and zwitterionic phospholipids. Independent measurements of the number of ions adsorbed to sonicated PS vesicles were made with a calcium-sensitive electrode. The results agreed with the zeta potential results obtained with multilamellar vesicles. When membranes are formed at 20 degrees C in 0.1 M NaCl, the intrinsic 1:1 association constants of Ni, Co, Mn, Ba, Sr, Ca, and Mg with PS are 40, 28, 25, 20, 14, 12, and 8 M-1, respectively.     

Adsorption of uni and divalent cations to planar bilayer membranes containing sulphatides or phosphatidylserine

It has been postulated that sulphatides may be the K+ binding site of the sodium pump. In order to test this hypothesis we studied the binding of K+ to bilayer membranes containing sulphatides or phosphatidylserine. The adsorption constants of Na+, K+ and Ca2+ to planar bilayers containing these acidic lipids were determined from changes in the electrostatic potential at the membrane surface. Our results indicate that univalent cations adsorb weakly to both lipids and Ca2+ binds more strongly. The sequence of ion binding was Ca2+ greater than Na+ greater than K+. These results indicate that K+ does not bind specifically to sulphatides or phosphatidylserine and rule out the proposal that sulphatides by themselves provide the K+ binding site of the sodium pump.     

Discrimination between monovalent and divalent cations by hydrophobic solvent-saturated membranes containing fixed negative charges

Summary Cellulose acetate-nitrate filters were saturated with hydrophobic solvent and interposed between various aqueous solutions. The membranes thus formed are cation permselective. The discrimination between a monovalent cation such as K⁺ and the alkaline earth group divalent cations is very sharp. The discrimination ratio is at least a few thousand times in favor of the monovalent cation. A major part of this discrimination is caused by the very low mobility of the divalent cation within the membrane compared with that of the monovalent cation. The remainder of the discrimination is caused by the selectivity of the membranes which prefer monovalent to divalent cations. There is a clear discrepancy between Ba⁺⁺ diffusibility and mobility within, the membrane. This implies that Ba⁺⁺ may move within the hydrophobic membrane as a neutral complex. Some similarity with natural biological membranes is indicated.     

Adsorption of monovalent and divalent cations by phospholipid membranes. The monomer-dimer problem.

A generalization of the Stern theory is derived to treat the simultaneous adsorption of monovalent cations and divalent cations by single-component phospholipid membranes, where the ion:phospholipid binding stoichiometries are 1:1 for the monovalent cations and 1:1 and/or 1:2 for the divalent cations. This study treats both the situation in which the monovalent and divalent cations compete for membrane binding sites and that in which they do not compete. The general formalism of the screening/binding problem is reviewed, and it is shown how the adsorption problem can be isolated from the electrostatics. The statistical mechanics of mixed 1:1- and 1:2-stoichiometric adsorption (the monomer-dimer problem) is treated, and the problem of simultaneous 1:1 and 1:2 binding is solved. A simple expression for this solution, given in the Bethe approximation, is combined with the electrostatics to yield an adsorption isotherm encompassing both 1:1 monovalent-cation, and 1:1 and 1:2 divalent-cation, binding to charged membranes. A comparison with the simplified treatment of previous authors is made and the significance of their assumptions clarified in light of the present result. The present and previous treatments are plotted for a representative case of Na+ and Ca++ binding to a phosphatidylserine membrane. Criteria are established to permit unambiguous experimental testing of the present vs. previous treatments.     

Interaction of gentamicin and spermine with bilayer membranes containing negatively charged phospholipids

We measured the electrophoretic mobility of multilamellar phospholipid vesicles, the 31P NMR spectra of both sonicated and multilamellar vesicles, and the conductance of planar bilayer membranes to study the binding of spermine and gentamicin to membranes. Spermine and gentamicin do not bind significantly to the zwitterionic lipid phosphatidylcholine. We measured the concentrations of gentamicin and spermine that reverse the charge on vesicles formed from a mixture of phosphatidylcholine and either phosphatidylserine or phosphatidylinositol. From these measurements, we determined that the intrinsic association constants of the cations with these negative lipids are all about 10 M-1. This value is orders of magnitude lower than the apparent binding constants reported in the literature by other groups because the negative electrostatic surface potential of the membranes and the resultant accumulation of these cations in the aqueous diffuse double layer adjacent to the membranes have not been explicitly considered in previous studies. Our main conclusion is that the Gouy-Chapman-Stern theory of the aqueous diffuse double layer can describe surprisingly well the interaction of gentamicin and spermine with bilayer membranes formed in a 0.1 M NaCl solution if the negative phospholipids constitute less than 50% of the membrane. Thus, the theory should be useful for describing the interactions of these cations with the bilayer component of biological membranes, which typically contain less than 50% negative lipids. For example, our results support the suggestion of Sastrasinh et al. [Sastrasinh, M., Krauss, T. C., Weinberg, J. M., & Humes, H. D. (1982) J. Pharmacol. Exp. Ther. 222, 350-358] that phosphatidylinositol is the major binding site for gentamicin in renal brush border membranes.     

The adsorption of divalent cations to phosphatidylcholine bilayer membranes

Electrophoretic mobility and ³¹P NMR measurements were combined to test whether the combination of the Henry, Boltzmann and Grahame equations is capable of describing the adsorption of divalent cations to phosphatidylcholine membranes. Cobalt was chosen for this study because, of all the common divalent cations, its effects on the ³¹P NMR spectrum of phosphatidylcholine membranes are easiest to interpret. Both the ³¹P NMR data on the adsorption of cobalt and the zeta potential data calculated from the electrophoretic mobility in the presence of cobalt are well described by the combination of these three equations. Electrophoretic mobility measurements were also performed with a number of other divalent cations and the zeta potentials were, in all cases, well described by the combination of these three equations. The binding deduced from such measurements decreases in the sequence: Mn²⁺, Mg²⁺, Ca²⁺, Co²⁺, Ni²⁺, Sr²⁺, Ba²⁺. If we assume that a lipid molecule occupies an area of 60 Ų and that there is a 1: 1 stoichiometry for the binding of the divalent ions to phosphatidylcholine, the dissociation constants are, respectively: 0.3, 1.0, 1.0, 1.2, 1.2, 2.8, 3.6 M.     

The adsorption of divalent cations to phosphatidylglycerol bilayer membranes

The ability of the Stern equation to describe the adsorption of divalent cations to phosphatidylglycerol membranes was tested by combining ³¹P-NMR and electrophoretic mobility measurements. In 0.1 M sodium chloride both the ³¹P-NMR and the zeta potential data are well described by the Stern equation. ³¹P-NMR and ¹³C-NMR results indicate that cobalt forms inner-sphere complexes only with the phosphate group of phosphatidylglycerol molecules and that a substantial fraction of the adsorbed cobalt ions form outer-sphere complexes. Evidence is presented that suggests the alkaline earth cations also bind to phospholipids mainly by forming outer sphere complexes. Electrophoretic mobility measurements were performed with several different divalent cations. In all cases the zeta potentials in 0.1 M sodium chloride were well described by the Stern equation. The intrinsic 1 : 1 association constants (M^?1) for the phosphatidylglycerol complexes decreased in the sequence: Mn²⁺, 11.5; Ca²⁺, 8.5; Ni²⁺, 7.5; Co²⁺, 6.5; Mg²⁺, 6.0; Ba²⁺, 5.5 and Sr²⁺, 5.0.     

Interactions of divalent cations with phosphatidylserine bilayer membranes

The interaction of divalent cations with a homologous series of diacylphosphatidylserines (diacyl-PS) has been studied by differential scanning calorimetry and X-ray diffraction. Hydrated di-C14-PS (DMPS) exhibits a gel leads to liquid-crystal bilayer transition at 39 degrees C (delta H = 7.2 kcal/mol of DMPS). With increasing MgCl2 concentration, progressive conversion to a phase exhibiting a high melting (98 degrees C), high enthalpy (delta H congruent to 11.0 kcal/mol of DMPS) transition is observed. Similar behavior is observed for DMPS with increasing CaCl2 concentration. In this case, the high-temperature transition of the Ca2+-DMPS complex occurs at approximately 155 degrees C and is immediately followed by an exothermic transition probably associated with PS decomposition. For di-C12-, di-C14-, di-C16- (DPPS), and di-C18-PS, the transition temperatures of the Ca2+-PS complexes are in the range 151-155 degrees C; only di-C10-PS exhibits a significantly lower value, 142 degrees C. A different pattern of behavior is exhibited by DPPS in the presence of Sr2+ or Ba2+, with transitions in the range 70-80 degrees C being observed. X-ray diffraction of the Ca2+-PS complexes at 20 degrees C provides evidence of structural homology. All Ca2+-PS complexes exhibit bilayer structures, the bilayer periodicity increasing linearly from 35.0 A for di-C10-PS to 52.5 A for di-C18-PS. Wide-angle X-ray diffraction data indicate that hydrocarbon chain "crystallization" occurs on Ca2+-PS complex formation.(ABSTRACT TRUNCATED AT 250 WORDS)     

Permeabilities of composite membranes containing phospholipids

Two types of artificial membranes containing a phospholipid were prepared and their permeabilities were measured around the phase-transition temperature of the phospholipid. The permeability of the membranes to a hydrophobic solute was higher than to a hydrophilic solute, and showed an abrupt change at the phase-transition temperature of the phospholipid, similar to that in biomembranes and liposomes, caused by the fluidity change of the phospholipid at this temperature.     

Binding of peptides with basic residues to membranes containing acidic phospholipids.

There are clusters of basic amino acids on many cytoplasmic proteins that bind transiently to membranes (e.g., protein kinase C) as well as on the cytoplasmic domain of many intrinsic membrane proteins (e.g., glycophorin). To explore the possibility that these basic residues bind electrostatically to monovalent acidic lipids, we studied the binding of the peptides Lysn and Argn (n = 1-5) to bilayer membranes containing phosphatidylserine (PS) or phosphatidylglycerol (PG). We made electrophoretic mobility measurements using multilamellar vesicles, fluorescence and equilibrium binding measurements using large unilamellar vesicles, and surface potential measurements using monolayers. None of the peptides bound to vesicles formed from the zwitterionic lipid phosphatidylcholine (PC) but all bound to vesicles formed from PC/PS or PC/PG mixtures. None of the peptides exhibited specificity between PS and PG. Each lysine residue that was added to Lys2 decreased by one order of magnitude the concentration of peptide required to reverse the charge on the vesicle; equivalently it increased by one order of magnitude the binding affinity of the peptides for the PS vesicles. The simplest explanation is that each added lysine binds independently to a separate PS with a microscopic association constant of 10 M-1 or a free energy of approximately 1.4 kcal/mol. Similar, but not identical, results were obtained with the Argn peptides. A simple theoretical model combines the Gouy-Chapman theory (which accounts for the nonspecific electrostatic accumulation of the peptides in the aqueous diffuse double layer adjacent to the membrane) with mass action equations (which account for the binding of the peptides to greater than 1 PS). This model can account qualitatively for the dependence of binding on both the number of basic residues in the peptides and the mole fraction of PS in the membrane.     

Photoinitiated ion movements in bilayer membranes containing magnesium octaethylporphyrin.

A photocurrent produced by planar lipid bilayers containing Mg-octaethylporphyrin in the presence of oxygen has been investigated to determine if the current is due to movement of the MgOEP+ ion in the bilayer. Photoexcitation of the MgOEP is known to produce MgOEP+ in the bilayer when an electron acceptor is present. However, the aqueous electron acceptors ferricyanide and methyl viologen (MV+2) have opposite effects on the photocurrent. Ferricyanide decreases the photo current, even in the presence of oxygen, whereas methyl viologen increases the photocurrent, but only when oxygen is present. We attribute most of the photocurrent to the movement of superoxide anion. The difference in effect between ferricyanide and methyl viologen is attributed to the different rates of reduction of O2 by reduced MV+ (fast) vs. ferrocyanide (slow) and the known competition between ferricyanide and oxygen as the acceptor for the photoexcited porphyrin. It is inferred that most of the MgOEP is localized in the polar region of the lipid bilayer. Addition of ferrocyanide to the aqueous phase on one side of the bilayer, to trap MgOEP+ produced on the other side by MV+2, fails to increase the lifetime of the photovoltage. With a pH gradient across the bilayer, we observed only 5% of the photovoltage expected for the selective transport of H+ or OH- by MgOEP+. Thus, these measurements set the lower limit for the cross bilayer transit time of MgOEP+ or its charge in the range of 0.1-0.5 s.     

Elastic deformation and failure of lipid bilayer membranes containing cholesterol.

Giant bilayer vesicles were reconstituted from several lipids and lipid/cholesterol (CHOL) mixtures: stearolyloleoylphosphatidylcholine (SOPC), bovine sphingomyelin (BSM), diarachidonylphosphatidylcholine (DAPC), SOPC/CHOL, BSM/CHOL, DAPC/CHOL, and extracted red blood cell (RBC) lipids with native cholesterol. Single-walled vesicles were manipulated by micropipette suction and several membrane material properties were determined. The properties measured were the elastic area compressibility modulus K, the critical areal strain alpha c, and the tensile strength tau lys, from which the failure energy or membrane toughness Tf was calculated. The elastic area expansion moduli for these lipid and lipid/cholesterol bilayers ranged from 57 dyn/cm for DAPC to 1,734 dyn/cm for BSM/CHOL. The SOPC/CHOL series and RBC lipids had intermediate values. The results indicated that the presence of cholesterol is the single most influential factor in increasing bilayer cohesion, but only for lipids where both chains are saturated, or mono- or diunsaturated. Multiple unsaturation in both lipid chains inhibits the condensing effect of cholesterol in bilayers. The SOPC/CHOL system was studied in more detail. The area expansion modulus showed a nonlinear increase with increasing cholesterol concentration up to a constant plateau, indicating a saturation limit for cholesterol in the bilayer phase of approximately 55 mol% CHOL. The membrane compressibility was modeled by a property-averaging composite theory involving two bilayer components, namely, uncomplexed lipid and a lipid/cholesterol complex of stoichiometry 1/1.22. The area expansion modulus of this molecular composite membrane was evaluated by a combination of the expansion moduli of each component scaled by their area fractions in the bilayer. Bilayer toughness, which is the energy stored in the bilayer at failure, showed a maximum value at approximately 40 mol% CHOL. This breakdown energy was found to be only a fraction of the available thermal energy, implying that many molecules (approximately 50-100) may be involved in forming the defect structure that leads to failure. The area expansion modulus of extracted RBC lipids with native cholesterol was compared with recent measurements of intact RBC membrane compressibility. The natural membrane was also modeled as a simple composite made up to a compressible lipid/cholesterol matrix containing relatively incompressible transmembrane proteins. It appears that the interaction of incompressible proteins with surrounding lipid confers enhanced compressibility on the composite structure.     

Adsorption of polyvalent cations to bilayer membranes from negatively charged lipid: estimating the lipid accessibility in the case of complete binding

The effect of trivalent (Gd(3+) and Yb(3+)) and divalent (Be(2+) and Ca(2+)) cations on suspensions of multilamellar liposomes formed from brain PS and DMPS has been studied using microelectrophoresis and DSC techniques, respectively. The zeta potential values have been shown to strongly depend on the total lipid concentration in the suspension. At moderate concentrations of the polyvalent cations, the total cation concentration exceeds the bulk one several times due to adsorption of cations to the liposomes. A modification of the Gouy-Chapman-Stern theory in the case of unknown bulk concentration of the polyvalent cation is presented. An intrinsic association constant for Be(2+) ions was evaluated to be about K(2) approximately 50 M(-1). The algorithm for estimating the concentrations of the accessible (to exogenously added polyvalent cations) lipid-binding sites is described. These values are consistent with the subsurface concentrations of the polyvalent cations, which monotonously increase with the total concentration of the polyvalent cations. The calculated lipid accessibilities are shown to be in accordance with the DSC data.     

Predicting stability of DNA duplexes in solutions containing magnesium and monovalent cations

Accurate predictions of DNA stability in physiological and enzyme buffers are important for the design of many biological and biochemical assays. We therefore investigated the effects of magnesium, potassium, sodium, Tris ions, and deoxynucleoside triphosphates on melting profiles of duplex DNA oligomers and collected large melting data sets. An empirical correction function was developed that predicts melting temperatures, transition enthalpies, entropies, and free energies in buffers containing magnesium and monovalent cations. The new correction function significantly improves the accuracy of predictions and accounts for ion concentration, G-C base pair content, and length of the oligonucleotides. The competitive effects of potassium and magnesium ions were characterized. If the concentration ratio of [Mg (2+)] (0.5)/[Mon (+)] is less than 0.22 M (-1/2), monovalent ions (K (+), Na (+)) are dominant. Effects of magnesium ions dominate and determine duplex stability at higher ratios. Typical reaction conditions for PCR and DNA sequencing (1.5-5 mM magnesium and 20-100 mM monovalent cations) fall within this range. Conditions were identified where monovalent and divalent cations compete and their stability effects are more complex. When duplexes denature, some of the Mg (2+) ions associated with the DNA are released. The number of released magnesium ions per phosphate charge is sequence dependent and decreases surprisingly with increasing oligonucleotide length.     

Ionic permeation of lipid bilayer membranes mediated by a neutral,noncyclic Li+-selective carrier having imide and ether ligands. I. Selectivity among monovalent cations

Summary We have found that Simon's neutral, noncyclic, Li⁺-selective complexone, which has imide and ether ligands, renders lipid bilayer membranes selectively permeable to certain cations and anions. The present paper characterizes the ability of this molecule to carry monovalent cations; and we show it to be most selective for Li⁺ among the alkali cations, the first reconstitution of Li⁺-selective permeation in lipid bilayer membranes. This complexone acts as an equilibrium-domain carrier for Ag⁺> Li⁺>Tl⁺>Na⁺>NH ₄ ⁺ >Rb⁺>Cs⁺ over a wide range of experimental conditions. The major type of membrane-permeating species formed is a 21 carrier/cation complex dominant except at the lowest salt and carrier concentrations where a 11 carrier/cation, with a similar selectivity sequence, can be detected. Among the groupIa cations the selectivity sequence in bilayers, Li⁺>Na⁺>K⁺>Rb⁺>Cs⁺, is similar to that previously found for this molecule in thick solvent-polymer membrane electrodes. We find this carrier to be more selective to Ag⁺ than to any other monovalent cation yet studied. This high Ag⁺ selectivity is used, together with the dependence of the selectivity on the nature of the N-amide substitutents, to argue that the imide oxygens play a major role as ligands.     

Selectivity of edta-modified calcium channels to monovalent cations

The characteristics of slow inward sodium currents arising in response to membrane depolarization were studied in experiments on isolated dialyzed neurons of the snailHelix pomatia when the calcium-chelating agent EDTA was added to the calcium-free external solution. Values of the relative permeability of the corresponding ionic channels, determined from the shift of the equilibrium potential, were: P_Na+:P_Li+: +=1.00:0.80:0.55:0.21. The ratio between these values for "fast" sodium channels was 1.00:1.04:0.44:0.19. The induced sodium current was blocked by D-600 and nifedipine, which block calcium channels, more effectively than the calcium current of the same membrane (the corresponding dissociation constants were 10^–5 and 0.8·10^–5 mole/liter for the induced sodium current compared with 2.6·10^–5 and 2.3·10^–5 mole/liter for the calcium current). It is postulated on the basis of these data that the calcium channels have a principal selective filter similar to that of sodium channels, but also an additional binding site for bivalent cations, which prevents entry of monovalent cations into the channel. The addition of calcium-chelating agents to the calcium-free external solution liberates this site and thereby modifies the calcium channel into a sodium channel.A. A. Bogomolets Institute of Physiology, Academy of Sciences of the Ukrainian SSR, Kiev. Translated from Neirofiziologiya, Vol. 14, No. 5, pp. 491–498, September–October, 1982.     

Swelling of phospholipids by monovalent salt

Critical to biological processes such as membrane fusion and secretion, ion-lipid interactions at the membrane-water interface still raise many unanswered questions. Using reconstituted phosphatidylcholine membranes, we confirm here that multilamellar vesicles swell in salt solutions, a direct indication that salt modifies the interactions between neighboring membranes. By varying sample histories, and by comparing with data from ion carrier-containing bilayers, we eliminate the possibility that swelling is an equilibration artifact. Although both attractive and repulsive forces could be modified by salt, we show experimentally that swelling is driven primarily by weakening of the van der Waals attraction. To isolate the effect of salt on van der Waals interactions, we focus on high salt concentrations at which any possible electrostatic interactions are screened. By analysis of X-ray diffraction data, we show that salt does not alter membrane structure or bending rigidity, eliminating the possibility that repulsive fluctuation forces change with salt. By measuring changes in interbilayer separation with applied osmotic stress, we have determined, using the standard paradigm for bilayer interactions, that 1 M concentrations of KBr or KCl decrease the van der Waals strength by 50%. By weakening van der Waals attractions, salt increases energy barriers to membrane contact, possibly affecting cellular communication and biological signaling.     

A time-resolved fluorescence anisotropy study of bilayer membranes containing alpha-tocopherol

Rotational mobility in fluid phase dipalmitoylphosphatidylcholine unilamellar vesicles containing alpha-tocopherol has been studied by time-resolved anisotropy measurements of fluorescence from a diphenylhexatriene-phosphatidylcholine conjugate. The results are analysed using a simple wobbling-in-cone model. The diphenylhexatriene probe shows an increasing order parameter and more restricted wobbling with increasing alpha-tocopherol content of the membrane. The diffusional rate for wobbling was found not to change significantly.