Quantifying the effects of melittin on liposomes

Melittin, the soluble peptide of bee venom, has been demonstrated to induce lysis of phospholipid liposomes. We have investigated the dependence of the lytic activity of melittin on lipid composition. The lysis of liposomes, measured by following their mass and dimensions when immobilised on a solid substrate, was close to zero when the negatively charged lipids phosphatidyl glycerol or phosphatidyl serine were used as the phospholipid component of the liposome. Whilst there was significant binding of melittin to the liposomes, there was little net change in their diameter with melittin binding reversed upon salt injection. For the zwitterionic phosphatidyl choline the lytic ability of melittin is dependent on the degree of acyl chain unsaturation, with melittin able to induce lysis of liposomes in the liquid crystalline state, whilst those in the gel state showed strong resistance to lysis. By directly measuring the dimensions and mass changes of liposomes on exposure to melittin using Dual Polarisation Interferometry, rather than following the florescence of entrapped dyes we attained further information about the initial stages of melittin binding to liposomes.     

Oxidant membrane injury by the neutrophil myeloperoxidase system. II. Injury by stimulated neutrophils and protection by lipid-soluble antioxidants

The stimulated human neutrophil can damage a variety of target cells, and in some models, a mechanism involving secretion of myeloperoxidase and H2O2 has been demonstrated. We explored the characteristics of this cell-cell interaction by using neutrophils and our recently described liposome model target cell system. Exposure of 51Cr-labeled liposomes to phorbol myristate acetate-stimulated human neutrophils resulted in release of 25 to 30% of the radioactivity. 51Cr release was abrogated by omission of the neutrophils, the phorbol ester or halide (iodide), replacement of the phorbol by an inactive congener, or addition of azide, cyanide, or catalase. Neutrophils from patients with hereditary absence of myeloperoxidase (MPO) or a failure of H2O2 formation (chronic granulomatous disease) did not cause liposome lysis unless purified MPO or a source of H2O2, respectively, was added. These data indicate that 51Cr release from liposomes is a consequence of the secretion of MPO and H2O2, which combine with extracellular halides to form a membrane lytic system. The influence of liposome composition on injury was then examined, with a focus on physiologically relevant lipid soluble antioxidants. Liposomes containing either alpha-tocopherol (0.33 to 1.67% of molar fraction of lipid) or beta-carotene (1.67% of molar fraction of lipid) were markedly resistant to lysis by the cellfree MPO-H2O2-chloride system. When the major structural lipid phosphatidyl choline was replaced by dipalmitoyl phosphatidyl choline, a synthetic phospholipid with no oxidizable double bonds, the resultant liposomes were totally resistant to lysis by the MPO-H2O2-chloride system. The addition of iodide to this system (i.e., both chloride and iodide present) changed the pattern of protection dramatically in that alpha-tocopherol and beta-carotene were no longer protective and the resistance of dipalmitoyl phosphatidyl choline liposomes was partial rather than complete. In contrast to iodide, the addition of bromide or thiocyanate did not have a major effect on the protection by antioxidants. Finally, we demonstrated protection by alpha-tocopherol or dipalmitoyl phosphatidyl choline against liposome lysis by phorbol-activated neutrophils. These studies illustrate the use of model phospholipid membranes in the characterization of oxygen-dependent cell-mediated cytotoxicity. Activated neutrophils lyse liposome targets through a MPO-dependent mechanism. Target properties, especially the content of lipid-soluble antioxidants, have a marked influence on susceptibility to lysis.(ABSTRACT TRUNCATED AT 400 WORDS)     

Orientation of membrane-bound melittin studied by a combination of HPLC and liquid secondary ion mass spectrometry (LSIMS)

Combining two analytical techniques, HPLC and liquid secondary ion mass spectrometry, the orientation of liposomal membrane-bound melittin was analyzed through its trypsin-digested products. We found that trypsin can access all proteolytic sites of the membrane-bound melittin when the liposomes have no transmembrane potential, whereas the proteolytic site near the N terminus of melittin is blocked when the liposomes have a negative transmembrane potential. The results suggest that the negative transmembrane potential may induce the melittin molecules to insert into the membrane perpendicularly, whereas melittin lies flat on the membrane surface in the absence of a negative potential.     

蜂毒肽在磷脂膜上取向的高效液相色谱和质谱研究

利用脂质体模型系统研究了跨膜电位对蜂毒肽在磷脂磷上取向的影响,采用高效液相色谱与质谱技术相结合方法,分析了蜂毒肽与磷脂膜作用后的胰蛋白酶酶解产物,结果发现,当蜂毒肽分子与没有跨膜电位的脂质体结合后,它的所有酶切位点都能够酶有效切断,当蜂毒肽与带负向膜电全的脂质体结合后,其Ｎ端的一个酶切位点被封闭,从而说明跨膜电位造成了蜂毒肽在膜上的重瓣取向。     

Interaction of liposomes with erythrocytes in the process of their lysis]

Interaction of phosphatidyl choline liposomes with erythrocytes during spontaneous lysis of the latter has been studied. It is shown that hemoglobin which is released during the lysis of erythrocytes is found by liposomes which in their turn are absorbed on the external erythrocyte surface. In this case the binding of hemoglobin by liposomes takes place with a greater speed than its release during erythrocyte lysis and is accompanied by a change in its conformation. Possibilities of the microcalorimetry methods for studying the interaction of liposomes with erythrocytes under the conditions mentioned above are considered.     

Membrane fusion between liposomes composed of acidic phospholipids and neutral phospholipids induced by melittin: a differential scanning calorimetric study

Melittin-induced membrane fusion between neutral and acidic phospholipids was examined in liposome systems with a high-sensitivity differential scanning calorimeter. Membrane fusion could be detected by calorimetric measurement by observing thermograms of mixed liposomal lipids. The roles of hydrophobic and electrostatic interactions were investigated in membrane fusion induced by melittin. Melittin, a bee venom peptide, is composed of a hydrophobic region including hydrophobic amino acids and a positively charged region including basic amino acids. When phosphatidylcholine liposomes were prepared in the presence of melittin, reductions in the phase transition enthalpies were observed in the following order; dimyristoylphosphatidylcholine (DMPC) > dipalmitoylphosphatidylcholine (DPPC) > distearoylphosphatidylcholine (DSPC) > dielaidoylphosphatidylcholine (DEPC). The plase transition enthalpy of an acidic phospholipid, dipalmitoylphosphatidylserine (DPPS), was raised by melittin at low concentrations, then reduced at higher concentrations. DPPC liposomes prepared in melittin solution were fused with DPPS liposomes when the liposomal dispersions were mixed and incubated. Similar fusion was observed between dipalmitoylphosphatidylcholine and dimyristoylphosphatidic acid (DMPA) liposomes. These results indicate that a peptide including hydrophobic and basic regions can mediate membrane fusion between neutral and acidic liposomes by hydrophobic and electrostatic interactions.     

Leakage of internal markers from erythrocytes and lipid vesicles induced by melittin, gramicidin S and alamethicin: a comparative study.

The membrane-disruptive capacities of melittin, derivatised melittins, alamethicin and gramicidin S have been compared for the human erythrocyte membrane and lipid vesicles of three different compositions (phosphatidylcholine, 85% phosphatidylcholine/15% phosphatidylserine, and a lipid analogue of the outer leaflet of the human erythrocyte membrane). The sensitivity to ionic strength, divalent metal ions and polylysine of release of fluorescent markers from liposomes and of haemoglobin from intact erythrocytes has been assayed. Acetyl melittin was found to he more effective than melittin in lysing phosphatidylcholine and phosphatidylcholine/phosphatidylserine vesicles, somewhat less effective in the lipid analogue and markedly less effective in lysing erythrocytes. Succinyl melittin was non-haemolytic, but was able to lyse lipid vesicles at a high concentration. Ca2+ inhibited melittin haemolysis at high ionic strength (150 mM NaCl), but produced a more complex response of stimulation followed by inhibition at low ionic strength. In lipid vesicles, Ca2+ either stimulated melittin lysis or was ineffective. Zn2+ exerted effects similar to Ca2+ with lipid vesicles at approx. 10-fold lower concentration except that a weak inhibition was observed for the erythrocyte membrane lipid analogue at high ionic strength. Polylysine strongly inhibited haemolysis by melittin at low ionic strength, but was ineffective or stimulatory in lipid vesicle lysis. High phosphate concentration also inhibited melittin haemolysis, but again no corresponding effect could he found in any of the lipid vesicle systems. These disparities between effects of melittin on erythrocytes and lipid vesicles support the proposal that melittin-protein interactions are of consequence to its haemolytic action. Similar experiments were performed with gramicidin S and alamethicin in order to compare their lytic properties with those of melittin. It was found that each lysin exhibited its own individual pattern of sensitivity to lipid composition, ionic strength and inhibition by cations. It thus appears likely that the detailed molecular interactions responsible for lysis are significantly different for each of these three agents.     

Osmotic and pH transmembrane gradients control the lytic power of melittin.

Transmembrane osmotic gradients applied on large unilamellar 1-palmitoyl-2-oleoyl-phosphatidylcholine vesicles were used to modulate the potency of melittin to induce leakage. Melittin, an amphipathic peptide, changes the permeability of vesicles, as studied using the release of entrapped calcein, a fluorescent marker. A promotion of the ability of melittin to induce leakage was observed when a hyposomotic gradient (i.e., internal salt concentration higher than the external one) was imposed on the vesicles. It is proposed that structural perturbations caused by the osmotic pressure loosen the compactness of the outer leaflet, which facilitates the melittin-induced change in membrane permeability. Additionally, we have shown that this phenomenon is not due to enhanced binding of melittin to the vesicles using intrinsic fluorescence of the melittin tryptophan. Furthermore, we investigated the possibility of using a transmembrane pH gradient to control the lytic activity of melittin. The potency of melittin in inducing release is known to be inhibited by increased negative surface charge density. A transmembrane pH gradient causing an asymmetric distribution of unprotonated palmitic acid in the bilayer is shown to be an efficient way to modulate the lytic activity of melittin, without changing the overall lipid composition of the membrane. We demonstrate that the protective effect of negatively charged lipids is preserved for asymmetric membranes.     

Liposomes can function as targets for natural killer cytotoxic factor but not for tumor necrosis factor

NK cells exert their lytic action through the release of NK cytotoxic factors (NKCF) after stimulation by the bound target cell. NKCF may be related to granule-derived perforin/cytolysin on one hand and to the pleiotropic cytokine TNF on the other hand. In the present study, we show that NKCF can also lyse artificial lipid vesicles, as had been reported previously for cytotoxic granules and cytolysin. The lysis of large unilamellar vesicles was monitored by measuring the release of the encapsulated fluorescent dye carboxyfluorescein. NKCF-induced lysis was only observed with liposomes composed of a complex mixture of lipids including acidic phospholipids. No lysis could be demonstrated if the liposomes contained phosphatidylcholine as the only phospholipid, suggesting some kind of lipid specificity for the action of NKCF. A remarkable finding was that neither recombinant nor natural TNF were able to lyse large unilamellar vesicles, irrespective of their lipid composition, indicating different ways of interaction of NKCF and TNF with artificial (and presumably also biological) membranes.     

Active function of membrane receptors for enveloped viruses : I. Specific requirement for liposome-associated sialoglycolipids, but not sialoglycoproteins, to allow lysis of phospholipid vesicles by reconstituted Sendai viral envelopes

Phospholipid liposomes composed of phosphatidylcholine (PC) and cholesterol (chol), bearing the sialoglycoprotein glycophorin (GP), are able to effectively bind Sendai virus particles, but not to be lysed by them. Incorporation of gangliosides (gangl) into the above phospholipid vesicles (yielding liposomes composed of PC/chol/gangl/GP), although not increasing their ability to interact with Sendai virions, rendered them susceptible to the viral lytic activity. This was inferred from the ability of the virus to induce release of carboxyfluorescein (CF) upon interaction at 37 degrees C with liposomes composed of PC/chol/gangl/GP. Lysis of liposomes required the presence of the two viral envelope glycoproteins, namely the hemagglutinin/neuraminidase (HN) and the fusion (F) polypeptides, and was inhibited by phenylmethyl sulfonylfluoride (PMSF), dithiothreitol (DTT) and trypsin, showing that virus-induced lysis of PC/chol/gangl/GP liposomes reflects the fusogenic activity of the virus. Incubation of Sendai virus particles with liposomes containing the acidic phospholipid dicetylphosphate (DCP) but lacking sialic acid containing receptors, also resulted in release of the liposome content. Lysis of these liposomes was due to the activity of the viral HN glycoprotein, therefore not reflecting the natural viral fusogenic activity. Fluorescence dequenching studies, using fluorescently labeled reconstituted Sendai virus envelopes (RSVE), have shown that the viral envelopes are able to fuse with neutral, almost to the same extent, as with negatively charged liposomes. However, fusion with negatively charged liposomes, as opposed to fusion with neutral liposomes, was mediated by the viral HN glycoprotein and not by the viral fusion polypeptide.     

Modulation of melittin-induced lysis by surface charge density of membranes.

Phosphorus NMR spectroscopy was used to characterize the importance of electrostatic interactions in the lytic activity of melittin, a cationic peptide. The micellization induced by melittin has been characterized for several lipid mixtures composed of saturated phosphatidylcholine (PC) and a limited amount of charged lipid. For these systems, the thermal polymorphism is similar to the one observed for pure PC: small comicelles are stable in the gel phase and extended bilayers are formed in the liquid crystalline phase. Vesicle surface charge density influences strongly the micellization. Our results show that the presence of negatively charged lipids (phospholipid or unprotonated fatty acid) reduces the proportion of lysed vesicles. Conversely, the presence of positively charged lipids leads to a promotion of the lytic activity of the peptide. The modulation of the lytic effect is proposed to originate from the electrostatic interactions between the peptide and the bilayer surface. Attractive interactions anchor the peptide at the surface and, as a consequence, inhibit its lytic activity. Conversely, repulsive interactions favor the redistribution of melittin into the bilayer, causing enhanced lysis. A quantitative analysis of the interaction between melittin and negatively charged bilayers suggests that electroneutrality is reached at the surface, before micellization. The surface charge density of the lipid layer appears to be a determining factor for the lipid/peptide stoichiometry of the comicelles; a decrease in the lipid/peptide stoichiometry in the presence of negatively charged lipids appears to be a general consequence of the higher affinity of melittin for these membranes.     

The effect of cyclization of magainin 2 and melittin analogues on structure, function, and model membrane interactions: implication to their mode of action

The amphipathic alpha-helical structure is a common motif found in membrane binding polypeptides including cell lytic peptides, antimicrobial peptides, hormones, and signal sequences. Numerous studies have been undertaken to understand the driving forces for partitioning of amphipathic alpha-helical peptides into membranes, many of them based on the antimicrobial peptide magainin 2 and the non-cell-selective cytolytic peptide melittin, as paradigms. These studies emphasized the role of linearity in their mode of action. Here we synthesized and compared the structure, biological function, and interaction with model membranes of linear and cyclic analogues of these peptides. Cyclization altered the binding of melittin and magainin analogues to phospholipid membranes. However, at similar bound peptide:lipid molar ratios, both linear and cyclic analogues preserved their high potency to permeate membranes. Furthermore, the cyclic analogues preserved approximately 75% of the helical structure of the linear peptides when bound to membranes. Biological activity studies revealed that the cyclic melittin analogue had increased antibacterial activity but decreased hemolytic activity, whereas the cyclic magainin 2 analogue had a marked decrease in both antibacterial and hemolytic activities. The results indicate that the linearity of the peptides is not essential for the disruption of the target phospholipid membrane, but rather provides the means to reach it. In addition, interfering with the coil-helix transition by cyclization, while maintaining the same sequence of hydrophobic and positively charged amino acids, allows a separated evaluation of the hydrophobic and electrostatic contributions to binding of peptides to membranes.     

Interaction of melittin with model membranes: effect on the size and permeability of liposomes

The influence of melittin, a monomer devoid of the phospholipase activity, on the size and permeability of liposomes from egg lecithin (PC), dimyristoylphosphatidylcholine (DMPC) and dipalmitoylphosphatidylcholine (DPPC) has been investigated by the methods of fluorescence spectroscopy, quasi-elastic light scattering and freeze-fracture electron microscopy. While studying calcein release from liposomes under the influence of melittin it has been shown that binding of melittin with a bilayer is a fast process which depends on the concentration lipid: protein (Ri) ratio as well as on the phase state of the lipid. The lipids being in the liquid-crystalline forms (PC and DMPC) are characterized by a more rapid release of the dye-stuff from liposomes than DPPC vesicles being in gel state with the same Ri. Under the influence of different melittin concentrations heterogeneity of the system and its medium hydrodynamic size of particles at first increases (100 less than or equal to Ri less than 500) due to their fusion and then these parameters decrease to the initial values.     

The presence of PEG-lipids in liposomes does not reduce melittin binding but decreases melittin-induced leakage

Poly(ethyleneglycol) (PEG), anchored at the surface of liposomes via the conjugation to a lipid, is commonly used for increasing the liposome stability in the blood stream. In order to gain a better understanding of the protective properties of interfacial polymers, we have studied the binding of melittin to PEG-lipid-containing membranes as well as the melittin-induced efflux of a fluorescent marker from liposomes containing PEG-lipids. We examined the effect of the polymer size by using PEG with molecular weights of 2000 and 5000. In addition, we studied the role of the anchoring lipid by comparing PEG conjugated to phosphatidylethanolamine (PE) which results in a negatively charged PEG-PE, with PEG conjugated to ceramide (Cer) which provides the neutral PEG-Cer. Our results show that interfacial PEG does not prevent melittin adsorption onto the interface. In fact, PEG-PE promotes melittin binding, most likely because of attractive electrostatic interactions with the negative interfacial charge density of the PEG-PE-containing liposomes. However, PEG-lipids limit the lytic potential of melittin. The phenomenon is proposed to be associated with the change in the polymorphic tendencies of the liposome bilayers. The present findings reveal that the protective effect associated with interfacial hydrophilic polymers is not universal. Molecules like melittin can sense surface charges borne by PEG-lipids, and the influence of PEG-lipids on liposomal properties such as the polymorphic propensities may be involved in the so-called protective effect.     

The presence of PEG-lipids in liposomes does not reduce melittin binding but decreases melittin-induced leakage.

Poly(ethyleneglycol) (PEG), anchored at the surface of liposomes via the conjugation to a lipid, is commonly used for increasing the liposome stability in the blood stream. In order to gain a better understanding of the protective properties of interfacial polymers, we have studied the binding of melittin to PEG-lipid-containing membranes as well as the melittin-induced efflux of a fluorescent marker from liposomes containing PEG-lipids. We examined the effect of the polymer size by using PEG with molecular weights of 2000 and 5000. In addition, we studied the role of the anchoring lipid by comparing PEG conjugated to phosphatidylethanolamine (PE) which results in a negatively charged PEG-PE, with PEG conjugated to ceramide (Cer) which provides the neutral PEG-Cer. Our results show that interfacial PEG does not prevent melittin adsorption onto the interface. In fact, PEG-PE promotes melittin binding, most likely because of attractive electrostatic interactions with the negative interfacial charge density of the PEG-PE-containing liposomes. However, PEG-lipids limit the lytic potential of melittin. The phenomenon is proposed to be associated with the change in the polymorphic tendencies of the liposome bilayers. The present findings reveal that the protective effect associated with interfacial hydrophilic polymers is not universal. Molecules like melittin can sense surface charges borne by PEG-lipids, and the influence of PEG-lipids on liposomal properties such as the polymorphic propensities may be involved in the so-called protective effect.     

Melittin lysis of red cells

Summary This paper describes experiments designed to explore interactions between human red blood cell membranes and melittin, the main component of bee venom. We found that melittin binds to human red cell membranes suspended in isotonic NaCl at room temperature, with an apparent dissociation constant of 3×10^–8 m and maximum binding capacity of 1.8×10⁷ molecules/cell. When about 1% of the melittin binding sites are occupied, cell lysis can be observed, and progressive, further increases in the fraction of the total sites occupied lead to progressively greater lysis in a graded manner. 50% lysis occurs when there are about 2×10⁶ molecules bound to the cell membrane. For any particular extent of melittin binding, lysis proceeds rapidly during the first few minutes but then slows and stops so that no further lysis occurs after one hour of exposure of cells to melittin. The graded lysis of erythrocytes by melittin is due to complete lysis of some of the cells, since both the density and the hemoglobin content of surviving, intact cells in a suspension that has undergone graded melittin lysis are similar to the values observed in the same cells prior to the addition of melittin. The cells surviving graded melittin lysis have an increased Na and reduced K, proportional to the extent of occupation of the melittin binding sites. Like lysis, Na accumulation and K loss proceed rapidly during the first few minutes of exposure to melittin but then stops so that Na, K and hemoglobin content of the cells remain constant after the first hour. These kinetic characteristics of both lysis and cation movements suggest that melittin modifies the permeability of the red cell membrane only for the first few minutes after the start of the interaction. Direct observation of cells by Nomarsky optics revealed that they crenate, become swollen and lyse within 10 to 30 sec after these changes in morphology are first seen. Taken together, these results are consistent with the idea that melittin produces lysis of human red cells at room temperature by a colloid osmotic mechanism.     

Hemolytic potency and phospholipase activity of some bee and wasp venoms

1. The action of crude venoms of four aculeate species: Apis mellifera, Vespa crabro, Vespula germanica and Vespula vulgaris on human erythrocytes was investigated in order to determine the lytic and phospholipase activity of different aculeate venoms and their ability to induce red blood cell hemolysis. 2. Bee venom was the only extract to completely lyse red blood cells at the concentration of 2-3 micrograms/ml. 3. Phospholipase activity in all of the examined vespid venoms was similar and the highest value was recorded in V. germanica. 4. Vespid venoms exhibited phospholipase B activity, which is lacking in honeybee venom. 5. In all membrane phospholipids but lecithin, lysophospholipase activity of vespid venoms was 2-6 times lower than the relevant phospholipase activity. 6. The incubation of red blood cells with purified bee venom phospholipase A2 was not accompanied by lysis and, when supplemented with purified melittin, the increase of red blood cell lysis was approximately 30%.     

Phospholipases: melittin facilitation of bee venom phospholipase A2-catalyzed hydrolysis of unsonicated lecithin liposomes.

Melittin isolated from the venom of the common honey bee is a potent activator for bee venom phospholipase A₂-catalyzed hydrolysis of unsonicated liposomes of egg phosphatidyl choline. At 37 °C and pH 8, the rate of this enzymatic reaction is increased approximately 300-fold by the addition of 8 × 10^?5m melittin. The magnitude of facilitation of the phospholipase A₂ reaction is much greater than that previously reported by other workers for systems involving sonicated egg phosphatidyl choline liposomes or Escherichia coli membrane fragments as substrates. Melittin having lysines quantitatively modified through reaction with methyl acetimidate is as effective a potentiator of phospholipase A₂ activity as the unmodified material. The same result was obtained for melittin in which the single tryptophan residue was modified. Melittin modified by succinylation retained approximately 50% of its capacity to facilitate phospholipase A₂ activity. In contrast, a modified melittin in which the C-terminal four amino residues were removed, acetimidated des(23–26)melittin, is a very poor activator, as is a mixture of this peptide with the C-terminal tetrapeptide. In contrast to the results with egg lecithin liposomes, melittin has little influence on the susceptibility of monomolecular aqueous solutions of dihexanoylphosphatidyl choline to phospholipase A₂ attack.     

Naturally occurring antibodies to liposomes. II. Specificity and electrophoretic pattern of rabbit antibodies reacting with sphingomyelin-containing liposomes

Sera obtained from rabbits after immunization with a variety of unrelated antigens contain antibodies that induce complement- (C) mediated lysis of sphingomyelin-containing liposomes in the absence of the relevant antigen from the membrane. Absorption or inhibition with dimyristoyl-phosphatidyl choline-containing liposomes were less effective than with sphingomyelin-containing liposomes in decreasing or abolishing C-dependent lysis of target-liposomes. Phosphoryl choline chloride inhibited the C-dependent lysis mediated by these antibodies, but only when used in high molar excess and in the presence of low antibody concentrations. Purified anti-liposome antibodies displayed an isoelectric focusing pattern consistent with a polyclonal response. The findings confirm the antibody nature of the anti-liposome activity of rabbit sera and indicate that their predominant specificity is directed against conformations of the phospholipid molecule in which the polar (phosphoryl choline) group does not have a major contribution.     

Aluminum-induced lipid phase separation and membrane fusion does not require the presence of negatively charged phospholipids

The interaction of Aluminum with phosphatidyl serine lipid vesicles containing variable amounts of phosphatidyl ethanolamine, phosphatidyl choline and cholesterol has been studied by lipid phase separation monitored by fluorescence quenching. The interaction of Al3+ with neutral phospholipid membranes has also been investigated. Maximal lipid phase separation can be demonstrated in mixed phosphatidyl ethanolamine-cholesterol vesicles when using concentrations of aluminum between 87.5 and 125 microM. Millimolar concentrations of Ca2+, Mn2+, Cd2+ and Zn2+ were without any effect. Aluminum also induced fusion of phospholipid membranes monitored by resonance energy transfer between N-(7-nitro-2,1,3, benzoxadiazol-4 yl) phosphatidyl ethanolamine and N-(lissamine Rhodamine B-sulfonyl) phosphatidyl ethanolamine, either when containing low amounts of phosphatidyl serine (12.5%) or without any negatively charged phospholipid. Aluminum-induced fusion of liposomes was also monitored by the fluorescence of the terbium-dipicolinic acid complex (Tb-DPA3-) formed during fusion of vesicles containing either Tb-(citrate)6- complex or sodium salt of dipicolinic acid.