D. Papahadjopoulos

Phase transitions in phospholipid vesicles Fluorescence polarization and permeability measurements concerning the effect of temperature and cholesterol

Abstract We have studied the solid to liquid-crystalline phase transition of sonicated vesicles of dipalmitoylphosphatidylglycerol and dipalmitoylphosphatidylcholine. The transition was studied by both fluorescence polarization of perylene embedded in the vesicles, and by the efflux rate of trapped 22Na+. Fluorescence polarization generally decreases with temperature, showing an inflection in the region 32–42°C with a mid-point of approximately 37.5 °C. On the other hand, the perylene fluorescence intensity increases abruptly in this region. To explain this result, we have proposed that, for T c where T c is the transition temperature, perylene is excluded from the hydrocarbon interior of the membranes, whereas, T c this probe may be accommodated in the membrane interior to a large extent. The self-diffusion rates of 22Na+ through dipalmitoylphosphatidylglycerol vesicles exhibit a complex dependence on temperature. There is an initial large increase in diffusion rates (approximately 100-fold) between 30 and 38 °C, followed by a decrease (approximately 4-fold) between 38 and 48 °C. A monotonic increase is then observed at temperatures higher than 48 °C. The local maximum of 22Na+ self-diffusion rates at approximately 38 °C coincides with the mid-point of phase transition as detected by changes in fluorescence polarization of perylene with the same vesicles. Vesicles composed of dipalmitoylphosphatidylcholine show the same general behavior in terms of 22Na+ self-diffusion rates at different temperatures, except that the local maximum occurs at approximately 42 °C. The temperature dependence of the permeability and the appearance of a local maximum at the phase transition region could be explained in terms of a domain structure within the plane of the membranes. This explanation is based on the possibility that boundary regions between liquid and solid domains would exhibit relatively high permeability to 22Na+. Mixed vesicles composed of equimolar amounts of dipalmitoyl phospholipids and cholesterol show no abrupt changes in the temperature dependence of either perylene fluorescence polarization or 22Na+ diffusion rate measurements. This is taken to indicate the absence of agross phase transition in the presence of cholesterol.

Cochleate lipid cylinders: formation by fusion of unilamellar lipid vesicles

Freeze-fracture electron microscopy was used to study the morphological changes occurring following the addition of Ca-2+ to sonicated preparations of phosphatidylserine in aqueous NaCl buffer. Before the addition of Ca-2+, preparations contained only small (200-500 A diameter) spheroidal vesicles. After the addition of Ca-2+ (10 mM) and incubation for 1 h at 37 degrees C preparations contained only large (2000-10 000 A) apparently multilamellar structures many of which were cylindrical in shape. The lamellae in these cylinders appear to be folded in a spiral configuration. Addition of EDTA to these preparations produced large, closed, spherical, unilamellar vesicles. We suggest the name cochleate lipid cylinders for the spiral structures and propose that they are formed by fusion of unilamellar vesicles into large sheets which fold spirally to form cylinders.

Membrane fusion and molecular segregation in phospholipid vesicles

Abstract Fusion between vesicles prepared from individual or mixed phospholipid species was demonstrated by ultracentrifugation and gel-filtration techniques, electron microscopy and differential scanning calorimetry. Variation of the chemical composition of the vesicles permitted evaluation of the effect of surface charge, Ca2+, fluidity and the presence of cholesterol on the fusion reaction and the segregation of lipid species within fused vesicles. Extensive fusion occurred between negatively charged phosphatidylserine vesicles incubated in the presence of CaCl2 ( > 1 mM ) and in vesicles prepared from greater than 50% phosphatidylserine in phosphatidylcholine in the presence of CaCl2 ( > 4 mM ) and albumin (0.1 mg/ml). Neutral phosphatidylcholine vesicles showed only a limited capacity to fuse. Vesicles containing lipids that were in a liquid-crystalline state were more susceptible to fusion than vesicles composed of lipids that were in the solid phase at experimental temperatures. Incorporation of equimolar amounts of cholesterol into vesicles composed of lipids in a liquid-crystalline state suppressed their ability to fuse. Calorimetric measuremens revealed Ca2+ induced segregation of individual lipids to form separate domains within the vesicle membrane (phase separation). The relationship of fusion between vesicles and fusion occurring in natural membranes was discussed.

Studies on membrane fusion. III. The role of calcium-induced phase changes

The interaction of phosphatidylserine vesicles with Ca2+ and Mg2+ has been examined by several techniques to study the mechanism of membrane fusion. Data are presented on the effects of Ca2+ and Mg2+ on vesicle permeability, thermotropic phase transitions and morphology determined by differential scanning calorimetry, X-ray diffraction, and freeze-fracture electron microscopy. These data are discussed in relation to information concerning Ca2+ binding, charge neutralization, molecular packing, vesicle aggregation, phase transitions, phase separations and vesicle fusion. The results indicate that at Ca2+ concentrations of 1.0-2.0 mM, a highly cooperative phenomenon occurs which results in increased vesicle permeability, aggregation and fusion of the vesicles. Under these conditions the hydrocarbon chains of the lipid bilayers undergo a phase change from a fluid to a crystalline state. The aggregation of vesicles that is observed during fusion is not sufficient range of 2.0-5.0 mM induces aggregation of phosphatidylserine vesicles but no significant fusion nor a phase change. From the effect of variations in pH, temperature, Ca2+ and Mg2+ concentration on the fusion of vesicles, it is concluded that the key event leading to vesicle membrane fusion is the isothermic phase change induced by the bivalent metals. It is proposed that this phase change induces a transient destabilization of the bilayer membranes that become susceptible to fusion at domain boundaries.

Effects of local anesthetics on membrane properties I. Changes in the fluidity of phospholipid bilayers

The effect of the local anesthetic dibucaine on the solid to liquid-crystalline phase transition in phospholipid vesicles was studied by calorimetry and fluorescence polarization. The partition coefficient (greater than 3000) of dibucaine in the membranes of vesicles prepared from acidic phospholipids was more than 20 times higher than in neutral phospholipid membranes under the same conditions. Calorimetric measurements on vesicles prepared form acidic phospholipids (bovine brain phosphatidylserine; dipalmitoylphosphatidylglycerol) showed that dibucaine (1 with 10(-4) M) produced a significant reduction in the gel-liquid crystalline transition temperature (Tc). This fluidizing effect of dibucaine on acidic phospholipid membranes was even more marked in the presence of Ca2+. In contrast, dibucaine at the same concentration did not alter the Tc of neutral phospholipids (dipalmitoylphosphatidylcholine). Significant increase in the fluidity of neutral phospholipid membranes occurred only at higher dibucaine concentrations (2 with 10(-3) M). Measurements of the fluorescence polarization and lifetime of the probe, 1,6-diphenylhexatriene, in acidic phospholipid vesicles revealed that dibucaine (1 with 10(-4) M) caused an increase in the probe rotation rate indicating an increase in the fluidity of the phospholipid membranes. A good correlation was obtained between fluorescence polarization data on dibucaine-induced changes in membrane fluidity and calorimetric measurements on vesicles of the same type.

Studies on membrane fusion. II. Induction of fusion in pure phospholipid membranes by calcium ions and other divalent metals

The effect of divalent metals on the interaction and mixing of membrane components in vesicles prepared from acidic phospholipids has been examined using freeze-fracture electron microscopy and differential scanning calorimetry. Ca2+, and to a certain extent Mg2+, induce extensive mixing of vesicle membrane components and drastic structural rearrangements to form new membranous structures. In contrast to the mixing of vesicle membrane components in the absence of Ca2+ described in the accompanying paper which occurs via diffusion of lipid molecules between vesicles, mixing of membrane components induced by Ca2+ or Mg2+ results from true fusion of entire vesicles. There appears to be a threshold concentration at which Ca2+ and Mg2+ become effective in inducing vesicle fusion and the threshold concentration varies for different acidic phospholipid species. Different phospholipids also vary markedly in their relative responsiveness to Ca2+ and Mg2+, with certain phospholipids being much more susceptible to fusion by Ca2+ than Mg2+. Vesicle fusion induced by divalent cations also requires that the lipids of the interacting membranes be in a fluid state (T greater than Tc). Fusion of vesicle membranes by Ca2+ and Mg2+ does not appear to be due to simple electrostatic charge neutralization. Rather the action of these cations in inducing fusion is related to their ability to induce isothermal phase transitions and phase separations in phospholipid membranes. It is suggested that under these conditions membranes become transiently susceptible to fusion as a result of changes in molecular packing and creation of new phase boundaries induced by Ca2+ (or Mg2+).

Studies on membrane fusion. 1. Interactions of pure phospholipid membranes and the effect of myristic acid, lysolecithin, proteins and dimethylsulfoxide

The interaction and mixing of membrane components in sonicated unilamellar vesicles and also non-sonicated multilamellar vesicles prepared from highly purified phospholipids suspended in NaCl solutions has been examined. Electron microscopy and differential scanning calorimetry were used to characterize the extent and kinetics of mixing of membrane components between different vesicle populations. No appreciable fusion was detected between populations of non-sonicated phospholipid vesicles incubated in aqueous salt (NaCl) solutions. Mixing of vesicle membrane components via diffusion of phospholipid molecules between vesicles was observed in populations of negatively charged phosphatidylglycerol vesicles but similar exchange diffusion was not detected in populations of neutral phosphatidylcholine vesicles. Incubation of sonicated vesicle populations at temperatures close to or above the phospholipid transition temperature resulted in an increase in vesicle size and mixing of vesicle membrane components as determined by a gradual change in the thermotropic properties of the mixed vesicle population. The interaction of purified phospholipid vesicles was also examined in the presence of myristic acid and lysolecithin. Our results indicate that while these agents enhance mixing of vesicle membrane components, in most cases mixing probably proceeds via diffusion of phospholipid molecules rather than by fusion of entire vesicles. Increased mixing of vesicle membrane components was also produced when vesicles were prepared containing a purified hydrophobic protein (myelin proteolipid apoprotein) or were incubated in the presence of dimethylsulfoxide. In these two systems, however, the evidence suggests that mixing of membrane components results from the fusion of entire vesicles.

Lipid vesicles as carriers for introducing biologically active materials into cells.

Publisher Summary Lipid vesicles have been used extensively over the past 10 years as model membranes from which to obtain information on the physicochemical organization of lipid bilayers and to study the interaction of various membrane-active drugs with lipids. More recently, experiments in several laboratories have shown that mammalian cells in vitro and in vivo will incorporate large numbers of lipid vesicles without cytotoxic effects. Since a wide variety of materials can be entrapped inside vesicles and various glycoproteins and glycolipids can be inserted into the vesicle membrane, cellular uptake of vesicles of defined composition offers a potential method for modifying cellular composition and for introducing nonpermeable biologically active materials into cells. This chapter discusses current methods for the production of different types of lipid vesicles and discusses the mechanism by which vesicles are incorporated into cells. Recent work on the use of lipid vesicles as “carriers” to introduce biologically active materials into cells in vitro and in vivo is also reviewed, and a brief outline is presented on the feasibility of developing methods in which vesicles could be targeted to specific parts of the cell and to particular cell types.

Effects of local anesthetics on membrane properties II. Enhancement of the susceptibility of mammalian cells to agglutination by plant lectins

Treatment of untransformed mouse and hamster cells with the tertiary amine local anesthetics dibucaine, tetracaine and procaine increases their susceptibility to agglutination by low doses of the plant lectin concanavalin A. Agglutination of anesthetic-treated untransformed cells by low doses of concanavalin A is accompanied by redistribution of concanavalin A receptors on the cell surface to form patches, similar to that occurring in spontaneous agglutination of virus-transformed cells by concanavalin A. Immunofluorescence and freeze-fracture electronmicroscopic observations indicate that local anesthetics per se do not induce this redistribution of concanavalin A receptors but modify the plasma membrane so that receptor redistribution is facilitated on binding of concanavalin A to the cell surface. Fluorescence polarization measurements on the rotational freedom of the membrane-associated probe, diphenylhexatriene, indicate that local anesthetics produce a small increase in the fluidity of membrane lipids. Spontaneous agglutination of transformed cells by low doses of concanavalin A is inhibited by colchicine and vinblastine but these alkaloids have no effect on concanavalin A agglutination of anesthetic-treated cells. Evidence is presented which suggests that local anesthetics may impair membrane peripheral proteins sensitive to colchicine (microtubules) and cytochalasin-B (microfilaments). Combined treatment of untransformed 3T3 cells with colchicine and cytochalasin B mimics the effect of local anesthetics in enhancing susceptibility to agglutination by low doses of concanavalin A. A hypothesis is presented on the respective roles of colchicine-sensitive and cytochalasin B-sensitive peripheral membrane proteins in controlling the topographical distribution of lectin receptors on the cell surface.

Cellular uptake of cyclic AMP captured within phospholipid vehicles and effect on cell-growth behaviour

Cyclic 3′,5′-adenosine monophosphate (cyclic AMP) was successfully incorporated within unilamellar phospholipid vesicles. Treatment of 3T3 and SV40-transformed 3T3 cells (SV3T3 cells) with vesicles containing cyclic AMP resulted in significant intracellular incorporation of vesicle-associated cyclic AMP and alteration of cell-growth behavior. Vesicles containing cyclic AMP caused a significant reduction in the growth rate of 3T3 and SV3T3 cells and effectively inhibited growth stimulation in stationary 3T3 cell cultures treated with insulin, serum and proteolytic enzymes. These alterations in cell-growth behaviour were produced by vesicle cyclic AMP at concentrations as low as 10−7 M. Similar modification of cell growth by addition of cyclic AMP or dibutyryl-cyclic AMP to the culture medium was achieved only at concentrations as high as 10−4 M plus 10−3 M theophylline. The ability of cyclic AMP trapped within vesicles to modify cell behaviour was influenced by the lipid composition of the vesicles. Cyclic AMP-containing vesicles composed of lipids that were in a “fluid” state of 37 °C produced marked growth inhibition while similar concentrations of cyclic AMP within vesicles prepared from “solid” lipids had no effect on cell proliferation. The influence of vesicle lipid composition on the mechanisms by which vesicles may be incorporated in to cells was discussed.