The lipidic particle as an intermediate structure in membrane fusion processes and bilayer to hexagonal HII transitions

Small unilamellar vesicles comprised of a mixture of phosphatidylethanolamine/phosphatidylcholine/cholesterol (3 : 1 : 2) fuse to form large multilamellar vesicles on increasing the temperature from 0 to 50°C. This event is associated with the appearance of lipidic particles at the fusion sites, consistent with a role as intermediary structures during the fusion process. Further, for phosphatidylcholine/cardiolipin (1 : 1) liposomes in the presence of Mn²⁺ a direct relationship between lipidic particles and the hexagonal (H_II) phase is demonstrated which suggests that lipidic particles can also occur as intermediaries between bilayer and hexagonal (H_II) structures.     

Cardiolipin and mitochondrial phosphatidylethanolamine have overlapping functions in mitochondrial fusion in Saccharomyces cerevisiae

The two non-bilayer forming mitochondrial phospholipids cardiolipin (CL) and phosphatidylethanolamine (PE) play crucial roles in maintaining mitochondrial morphology. We have shown previously that CL and PE have overlapping functions, and the loss of both is synthetically lethal. Because the lack of CL does not lead to defects in the mitochondrial network in Saccharomyces cerevisiae, we hypothesized that PE may compensate for CL in the maintenance of mitochondrial tubular morphology and fusion. To test this hypothesis, we constructed a conditional mutant crd1Δpsd1Δ containing null alleles of CRD1 (CL synthase) and PSD1 (mitochondrial phosphatidylserine decarboxylase), in which the wild type CRD1 gene is expressed on a plasmid under control of the TET(OFF) promoter. In the presence of tetracycline, the mutant exhibited highly fragmented mitochondria, loss of mitochondrial DNA, and reduced membrane potential, characteristic of fusion mutants. Deletion of DNM1, required for mitochondrial fission, restored the tubular mitochondrial morphology. Loss of CL and mitochondrial PE led to reduced levels of small and large isoforms of the fusion protein Mgm1p, possibly accounting for the fusion defect. Taken together, these data demonstrate for the first time in vivo that CL and mitochondrial PE are required to maintain tubular mitochondrial morphology and have overlapping functions in mitochondrial fusion.     

The integrity of proteoliposomes adsorbed on a biosurface

Proteoliposomes encapsulating [¹⁴C]glucose have been prepared from a mixture of dipalmitoylphosphatidylcholine and phosphatidylinositol by sonication (SUV) and reverse phase evaporation (REV) and conjugated with wheat germ agglutinin (WGA). The proteoliposomes were characterised in terms of size and composition and covered a range of size (weight-average diameter) from approximately 60–330 nm and surface-bound WGA (weight-average number of protein molecules per liposome) from approximately 70 to 3000. Methods have been developed for assessing the extent of adsorption and integrity of the proteoliposomes when targeted to glycophorin A-coated microtitre wells. From the amount of [¹⁴C]glucose released by detergent disruption from the adsorbed proteoliposomes it is found that the extent of adsorption increases with proteoliposome size and WGA conjugation and that the integrity of the proteoliposomes remains intact on adsorption. The results can be explained in terms of monolayer coverage of the surface with preferential adsorption of larger proteoliposomes from the size distribution.     

Membrane fusion of secretory vesicles and liposomes Two different types of fusion

Secretory vesicles isolated from adrenal medulla were found to fuse in vitro in response to incubation with Ca²⁺. Intervesicular fusion was detected by electron microscopy and was indicated by the appearance of twinned vesicles in freeze-fractured suspensions of vesicles and in thin-sectioned pellet. Two types of fusion could be distinguished: Type I, occurring between 10^?7 M and 10^?4 M Ca²⁺, was specific for Ca²⁺, was inhibited by other divalent cations and was abolished by pretreatment of vesicles with glutaraldehyde, neuraminidase or trypsin. Fusion type I was linear with temperature. A second type of intervesicular fusion was elicited by Ca²⁺ in concentrations higher than 2.5 mM and was morphologically characterized by multiple fusions of secretory vesicles. This type of fusion was found to be similar to fusion of liposomes prepared from the membrane lipids of adrenal medullary secretory vesicles: Ca²⁺ could be replaced by other divalent cations, the effect of different divalent cations was additive and pretreatments attacking membrane proteins were ineffective. Fusion type II of intact secretory vesicles as well as liposome fusion was discontinuous with temperature. Liposome fusion could be detected within 35 ms and persisted for 180 min. Using liposomes containing defined Ca²⁺ concentrations we have not found a major influence of Ca²⁺ asymmetry on fusion. Incorporation of the ganglioside GM₃, which is present in the membranes of intact adrenal medullary secretory vesicles did not change the properties of liposomes fusion. Using a Ca²⁺-selective electrode we have identified in secretory vesicle membranes both high affinity binding sites for Ca²⁺ ($\text{K}{}_{\mathrm{<mtext>d</mtext>}}\text{= 1.6 · 10}{}^{\mathrm{?6}}\text{M}$) and low affinity sites ($\text{K}{}_{\mathrm{<mtext>d</mtext>}}\text{= 1.2 · 10}{}^{\mathrm{?4}}\text{M}$).     

Immobilization of histidine-tagged proteins on monodisperse metallochelation liposomes: Preparation and study of their structure

Liposomes represent a biocompatible platform for the construction of self-assembling proteoliposomes using nickel or zinc metallochelation. Potential applications of such structures consist in the development of new biocompatible vaccination nanoparticles and drug delivery nanoparticle systems. Here, we describe the design and construction of a flow-through ultrafiltration cell suitable for the preparation of monodisperse liposomes enabled for metallochelation and, hence, the formation of proteoliposomes. The linkage of the cell with a fast protein liquid chromatography system facilitates automation of the procedure, which fits the criteria for upscaling. Proof-of-concept experiments are performed using a mixture of egg phosphatidyl choline and nickel-chelating lipid DOGS–NTA–Ni (1,2-dioleoyl-sn-glycero-3-{[N(5-amino-1-carboxypentyl)iminodiacetic acid]succinyl}(nickel salt)) to formulate proteoliposomes with proteins attached by metallochelation, including histidine (His)-tagged recombinant green fluorescent protein and rgp120 (derived from HIV-1 Env). These model proteoliposomes are characterized by gel permeation chromatography and by dynamic light scattering. Transmission electron microscopy and immunogold staining are used to characterize surface-bound proteins, revealing the tendency of rgp120 to form microdomains on liposome surfaces. These microdomains possess a two-dimensional crystal-like structure that is seen more precisely by atomic force microscopy.     

Docosahexaenoic acid-containing phosphatidylethanolamine in the external layer of liposomes protects docosahexaenoic acid from 2,2'-azobis(2-aminopropane)dihydrochloride-mediated lipid peroxidation

We have proposed that incorporation of docosahexaenoic acid (DHA) into phosphatidylethanolamine (PE) might enhance resistance to lipid peroxidation in vivo. In this study, we examined the relationship between the transbilayer distribution of PE and the oxidative stability of DHA in PE. Liposomes composed of a phospholipid mixture were used as models for biological membranes. To modulate the transbilayer distribution of PE obtained from the liver of rats fed DHA (PE-DHA), we used phosphatidylcholine (PC) with two types of acyl chain region: dipalmitoyl (PC16:0) or dioleoyl (PC18:1). The proportion of PE-DHA in the liposomal external layer was significantly higher in liposomes containing PC18:1 than in those containing PC16:0. This tendency was more pronounced in liposomes extruded using a polycarbonate filter with smaller pore sizes. Additionally, PE-DHA in the external layer of liposomes prepared using a filter with smaller pore sizes could protect DHA itself from 2,2(')-azobis(2-aminopropane)dihydrochloride-mediated lipid peroxidation.     

Effect of phophatidylcholine and phosphatidylethanolamine enrichment on the structure and function of yeast membrane

The phospholipid composition of yeast plasma membrane was manipulated by two different methods: (i) by using two auxotrophic strains KA101 (cho1) and MC13 (Cho⁺) which required phospholipid bases for growth and (ii) by supplementing Saccharomyces cerevisiae (3059) cells with high concentration of choline or ethanolamine. It was possible to enrich the plasma membrane with phosphatidylcholine (PC) or phosphatidylethanolamine (PE) by both methods. The uptake of amino acids, e.g., glycine, glutamic acid, leucine, lysine methionine, phenylalanine, proline and serine, was significantly reduced in PC- or PE-enriched cells. However, the extent of reduction in transport was variable among different strains. A fluorescent probe, 1-anilino-8-naphthalene sulfonate (ANS), was used to monitor the structural changes induced by altered phospholipid composition. It was observed that the relative fluorescence intensity of bound ANS was decreased as a consequence of PC or PE enrichment. The decrease in fluorescence was probably associated with reduced number of available binding sites (n) and increased apparent dissociation constant (K_d). Furthermore, our results also suggest that a critical level of PE or PC is required for proper functioning of yeast membrane.     

Energized transport of potassium ions in the absence of valinomycin by cytochrome c oxidase-reconstituted vesicles

Valinomycin-independent energized uptake of K⁺ was observed in cytochrome c oxidase reconstituted proteoliposome. The rate of K⁺ influx was proportoinal to the magnitude of electron flux. The energized uptake of K⁺ was abolished by p-trifluoromethoxycarbonylcyanide phenylhydrazone or by nigericin. Using the safranine fluorescence technique, it was demonstrated that even in the absence of valinomycin, liposomes and proteoliposomes reconstituted with cytochrome c oxidase are able to discriminate between Na⁺ and K⁺ and show a preference for K⁺ in the presence of excess Na⁺.     

Influence of local and neutral anaesthetics on the polymorphic phase preferences of egg yolk phosphatidylethanolamine

(1) The polymorphic phase preferences of egg phosphatidylethanolamine have been examined in the presence of normal alcohols and alkanes of varying chain length, as well as charged amine anaesthetics. (2) It is shown that the charged anaesthetics, ethanol and butanol can stabilize a bilayer arrangement for egg phosphatidylethanolamine. In contrast, longer chain ($\text{C}\text{?6}$) normal alcohols and alkanes induce the hexagonal (H_II) phase. (3) The relative potency of local anaesthetics in vitro (chlorpromazine, dibucaine, tetracaine and procaine) is mirrored by their relative ability to stabilize bilayer structure for hydrated egg phosphatidylethanolamine. Further, the aqueous concentrations of anaesthetic required to affect phospholipid polymorphism is sensitive to the lipid composition. For example, the inclusion of 20 mol% egg phosphatidylserine in egg phosphatidylethanolamine dispersions can reduce the aqueous concentrations of dibucaine required to induce appreciable bilayer stabilization effects from 5.0 mM to 0.5 mM. (4) It is suggested that the ability of amphipatic molecules such as anaesthetics to influence phosphatidylethanolamine polymorphism arises from their molecular shape. The possibility that anaesthetic molecules may exert their effects by virtue of this shape property is raised.     

The effect of pH on the oxygen kinetics of cytochrome c oxidase proteoliposomes

The effect of pH on the oxygen kinetics of cytochrome c oxidase incorporated into phospholipid vesicles is studied. The pH profiles of the oxygen kinetics of energized and deenergized oxidase vesicles are similar. An effect of pH on the slope of the reciprocal plot of rate against oxygen concentration is observed, and this may indicate that protons are involved in the rate limiting step of the reaction between oxygen and reduced oxidase. In contrast to the pH dependence of the oxygen kinetics, the binding of CO to the oxidase is not pH dependent.     

ATP-dependent calcium transport in membrane vesicles of the cyanobacterium, Anabaena variabilis

Transport of Ca²⁺ in membrane vesicles of the cyanobacterium Anabaena variabilis has been investigated. The light membranes previously shown to carry a Mg²⁺-dependent, Ca²⁺-stimulated ATPase (Lockau, W. and Pfeffer, S. (1982) Z. Naturforsch. 37C, 658–664) accumulate Ca²⁺ upon addition of ATP, whereas the (heavier) thylakoids do not. A stoichiometry of 0.3 Ca²⁺ taken up per ATP hydrolyzed has been determined from initial rates, which is considered to be an underestimation of the true stoichiometry of the pump. Calcium transport and Ca²⁺-stimulated ATPase activity are both sensitive to Na₃VO₄ (an inhibitor of ATPases forming a phosphorylated intermediate), show the same pH optimum and a comparable dependence on ATP concentration. Calcium transport is also supported by nucleoside triphosphates other than ATP, although at lower rates. Accumulation of calcium is abolished by an ionophore of divalent cations, ionophore A23187, but is resistant to ionophores of monovalent cations and to the inhibitor of F₁-F₀-type ATPases, N,N′-dicyclohexylcarbodiimide. It is concluded that the ATPase is a primary calcium pump.     

Molecular aspects of the bilayer stabilization induced by poly(l-lysines) of varying size in cardiolipin liposomes

The interaction between poly(l-lysines) of varying size with cardiolipin was investigated via binding assays, X-ray diffraction, freeze-fracture electron microscopy, and ³¹P- and ¹³C-NMR. Binding of polylysines to the lipid only occurred when three or more lysine residues were present per molecule. The strength of the binding was highly dependent on the polymerization degree, suggesting a cooperative interaction of the lysines within the polymer. Upon binding, a structural reorganization of the lipids takes place, resulting in a closely packed multilamellar system in which the polylysines are sandwiched in between subsequent bilayers. Acyl chain motion is reduced in these liquid-crystalline peptide-lipid complexes. From competition experiments with Ca²⁺ it could be concluded that when the affinity of the polylysine for cardiolipin was much larger than that of Ca²⁺, a lamellar polylysine-lipid complex was formed, irrespective of whether an excess of Ca²⁺ was added prior to or after the polypeptide. When the affinity of the polylysine for cardiolipin was less or of the same order as that of Ca²⁺, the lipid was organized in the hexagonal H_II phase in the presence of Ca²⁺. These results are discussed in the light of the peptide specificity of bilayer (de)stabilization in cardiolipin model membranes.     

Gramicidin promotes formation of the hexagonal HII phase in aqueous dispersions of phosphatidylethanolamine and phosphatidylcholine

It is shown by ³¹P-NMR and electron microscopy that gramicidin promotes the formation of the hexogonal H_II phase in aqueous dispersions of dielaidoylphosphatidylethanolamine and dioleoylphosphatidylethanolamine, when present in molar ratios of 1 : 200 and higher. In addition gramicidin also induces the hexogonal H_II phase in aqueous dispersions of dioleoylphosphatidylcholine, when present in molar ratios of 1 : 25 and higher.     

The involvement of parenchymal,Kupffer and endothelial liver cells in the hepatic uptake of intravenously injected liposomes effects of lanthanum and gadolinium salts

¹²⁵I-labeled albumin or poly(vinyl pyrrolidone) encapsulated in intermediate size multilamellar or unilamellar liposomes with 30–40% of cholesterol were injected intravenously into rats. In other experiments liposomes containing phosphatidyl[Me-¹⁴C]choline were injected. 1 h after injection parenchymal or non-parenchymal cells were isolated. Non-parenchymal cells were separated by elutriation centrifugation into a Kupffer cell fraction and an endothelial cell fraction. From the measurements of radioactivities in the various cell fractions it was concluded that the liposomes are almost exclusively taken up by the Kupffer cells. Endothelial cells did not contribute at all and hepatocytes only to a very low extent to total hepatic uptake of the ¹²⁵I-labels. Of the ¹⁴C-label, which orginates from the phosphatidylcholine moiety of the liposomes, much larger proportions were recovered in the hepatocytes. A time-dependence study suggested that besides the involvement of phosphatidylcholine exchange between liposomes and high density lipoprotein, a process of intercellular transfer of lipid label from Kupffer cells to the hepatocytes may be involved in this phenomenon. Lanthanum or gadolinium salts, which effectively block Kupffer cell activity, failed to accomplish an increase in the fraction of liposomal material recovered in the parenchymal cells. This is compatible with the notion that liposomes of the type used in these experiments have no, or at most very limited, access to the liver parenchyma following their intravenous administration to rats.     

Control of membrane fusion by phospholipid head groups II. The role of phosphatidylethanolamine in mixtures with phosphatidate and phosphatidylinositol

Membrane fusion induced by Ca²⁺ and Mg²⁺ in large unilamellar vesicles composed of mixtures of phosphatidylethanolamine with phosphatidate and phosphatidylinositol was studied by means of a fluorescence assay for the intermixing of internal aqueous contents of the vesicles. The threshold concentrations of Ca²⁺ or Mg²⁺ required for fusion increased only moderately when up to 80 mol% phosphatidylethanolamine was included with phosphatidate at pH 7.4, but no fusion could be detected in vesicles containing 70 mol% phosphatidylcholine even at high concentrations of Ca²⁺ or Mg²⁺. Phosphatidate-phosphatidylethanolamine (1 : 4) vesicles could be induced to fuse by 0.1 mM Ca²⁺ in the presence of a Mg²⁺ concentration which alone was insufficient for fusion. When equimolar amounts of phosphatidylethanolamine was included with phosphatidylinositol, the vesicles were susceptible to fusion by Ca²⁺, although pure phosphatidylinositol vesicles themselves merely aggregate and do not fuse (Sundler, R. and Papahadjopoulos, D. (1981) Biochim. Biophys. Acta 649, 743–750, accompanying paper). The role of phosphatidylethanolamine acyl chains, and hence the possible involvement of the bilayer-hexagonal (H_II) transition in membrane fusion, was examined by the temperature dependence of Ca²⁺-induced fusion in phosphatidylinositol-dimyristoylphosphatidylethanolamine (1 : 1) vesicles. Fusion was strictly dependent on the gel-liquid crystalline transition of the mixture and not on the phase behavior of the phosphatidylethanolamines. Comparable fusion rates were obtained for both egg yolk phosphatidylethanolamine and dimyristoylphosphatidylethanolamine at 50°C. As the dimyristoylphosphatidylethanolamine does not convert to a non-bilayer phase in this temperature range, we conclude that the bilayer-hexagonal transition is not necessary for membrane fusion. We propose that the dehydration characteristics of the phospholipids and their metal ion complexes are the critical factors determining fusion suceptibility of phospholipid membranes.     

Synthesis of N-hydroxysuccinimide esters of phosphatidylethanolamine and some properties of liposomes containing these derivatives

The N-hydroxysuccinimide (NHS) ester of N-suberyl-dimyristoylphosphatidylethanolamine (sub-DMPE) was synthesized by reaction of DMPE with disuccinimidyl suberate, and isolated by preparative plate chromatography. Liposomes, which contain NHS-sub-DMPE, can covalently bind compounds that possess a free amino group such as ε-dinitrophenyl-lysine. The extent of DNP-lysine binding is influenced by the time and temperature of incubation, the amount of NHS-sub-DMPE incorporated into the liposomes, and the initial concentration of DNP-lysine. Binding occurs as a consequence of the formation of a new dinitrophenylated compound which has been characterized. Although NHS-sub-DMPE is stable to storage in organic solvents, preformed liposomes rapidly lose their ability to bind DNP-lysine due to hydrolysis of the N-hydroxysuccinimide ester bond. These findings bear on the future applicability of liposomes, containing N-hydroxysuccinimide esters of PE, as illustrated by the preparation of immunogenic liposomes.     

Phosphatidylserine and phosphatidylethanolamine regulate the structure and function of FVIIa and its interaction with soluble tissue factor

Cell membranes have important functions in many steps of the blood coagulation cascade, including the activation of factor X (FX) by the factor VIIa (FVIIa)-tissue factor (TF) complex (extrinsic X_ase). FVIIa shares structural similarity with factor IXa (FIXa) and FXa. FIXa and FXa are regulated by binding to phosphatidylserine (PS)-containing membranes via their γ-carboxyglutamic acid-rich domain (Gla) and epidermal growth-factor (EGF) domains. Although FVIIa also has a Gla-rich region, its affinity for PS-containing membranes is much lower compared with that of FIXa and FXa. Research suggests that a more common endothelial cell lipid, phosphatidylethanolamine (PE), might augment the contribution of PS in FVIIa membrane-binding and proteolytic activity. We used soluble forms of PS and PE (1,2-dicaproyl-sn-glycero-3-phospho-l-serine (C6PS), 1,2-dicaproyl-sn-glycero-3-phospho-ethanolamine (C6PE)) to test the hypothesis that the two lipids bind to FVIIa jointly to promote FVIIa membrane binding and proteolytic activity. By equilibrium dialysis and tryptophan fluorescence, we found two sites on FVIIa that bound equally to C6PE and C6PS with K_d of ∼ 150–160 μM, however, deletion of Gla domain reduced the binding affinity. Binding of lipids occurred with greater affinity (K_d∼70–80 μM) when monitored by FVIIa proteolytic activity. Global fitting of all datasets indicated independent binding of two molecules of each lipid. The proteolytic activity of FVIIa increased by ∼50–100-fold in the presence of soluble TF (sTF) plus C6PS/C6PE. However, the proteolytic activity of Gla-deleted FVIIa in the presence of sTF was reduced drastically, suggesting the importance of Gla domain to maintain full proteolytic activity.     

Effect of microwave radiation on the permeability of carbonic anhydrase loaded unilamellar liposomes

The influence of 2.45 GHz microwave exposure (6 mW/g) on the diffusion processes in enzyme-loaded unilamellar liposomes as bioreactors was studied. The enzyme carbonic anhydrase (CA) was entrapped into cationic unilamellar vesicles. Previous kinetic experiments showed a very low self-diffusion rate of the substrate p-nitrophenyl acetate (PNPA) across intact liposome bilayer. A twofold increase in the diffusion rate of PNPA through the lipid bilayer was observed after 120 min of microwave radiation compared to temperature control samples. The microwave effect was time dependent. The enzyme activity, as a function of increased diffusion of PNPA, rises over 120 min from 22.3% to 80%. The increase in stearylamine concentration reduces the enzyme activity from 80% to 65% at 120 min. No enzyme leakage was observed. © 1994 Wiley-Liss, Inc.     

Retention of aqueous contents during divalent cation-induced fusion of phospholipid vesicles

The relative kinetics of intermixing and release of liposome aqueous contents during Ca²⁺-induced membrane fusion has been investigated. Fusion was monitored by the Tb-dipicolinic acid (DPA) fluorescence assay. Release was followed by the relief of self-quenching of carboxyfluorescein or by Tb fluorescence, with essentially identical results. Fusion of large unilamellar vesicles (LUV) made of phosphatidylserine (PS) in 100 mM NaCl (pH 7.4) at 25°C was initially non-leaky, whereas the fusion of small unilamellar vesicles (SUV) was accompanied by partial release of contents. After several rounds of fusion, the internal aqueous space of the vesicles collapsed. The rate of intermixing of lipids, measured by a resonance energy transfer assay, and the rate of coalescence of aqueous contents during fusion were similar over a range of Ca²⁺ concentrations. Most of the aqueous contents were retained after the fusion of SUV (PS) in 5 mM NaCl and 1 mM Ca²⁺. LUV made of a 1:1 mixture of Bacillus subtilis cardiolipin and dioleoylphosphatidylcholine went through about two rounds of fusion in the presence of Ca²⁺ at 10°C, with complete retention of contents. Similar results were obtained with vesicles composed of phosphatidate/PS/phosphatidylethanolamine/cholesterol (1:2:3:2) in the presence of Ca²⁺ and synexin at 25°C. These results emphasize the diversity of the relative kinetics of fusion and release in different phospholipid vesicle systems under various ionic conditions, and indicate that the initial events in the fusion of LUV are in general, non-leaky.