Calcium and magnesium dependence of phospholipase A2-catalyzed hydrolysis of phosphatidylcholine small unilamellar vesicles.

The Ca2+ requirement for lipid hydrolysis catalyzed by phospholipase A2 from Agkistrodon piscivorus piscivorus (App-D49) and porcine pancreas has been examined using small, unilamellar vesicles of dipalmitoylphosphatidylcholine (DPPC SUV). Hydrolysis was affected by product inhibition even at early times, and the extent of this inhibition depended on the concentration of divalent cations. The Ca2+ requirement for half-maximal rates of hydrolysis reflected, in part, this non-catalytic role of divalent cations. The presence of 10 mM Mg2+, a cation which does not support catalysis, reduced the Ca2+ required for half-maximal rates of hydrolysis from millimolar concentrations to 40 microM for App-D49. Since the dissociation constant of the enzyme for Ca2+ in solution is 2 mM, these results indicate a change in the interaction of the enzyme with Ca2+ under catalytic conditions. The kinetic dissociation constant of Ca2+ for the pancreatic enzyme was 20 microM which is substantially lower than the dissociation constant in solution, 0.35 mM. The similarity of apparent kinetic dissociation constants for these enzymes suggests that structurally similar features determine the affinity for Ca2+ under catalytic conditions. Evidence is presented that the affinity of phospholipase A2 for Ca2+ changes subsequent to the initial interaction of the enzyme with the substrate interface. However, the apparent Michaelis constant, KMapp, for App-D49, 0.03-0.06 mM, is independent of [Ca2+] and is about the same as the equilibrium dissociation constant for DPPC SUV, 0.14 mM. We thus suggest that KMapp is a steady-state constant.     

Spontaneous fusion of phosphatidylcholine small unilamellar vesicles in the fluid phase

Using a high-sensitivity differential scanning microcalorimeter capable of performing cooling scans, we have examined the phase behavior of small unilamellar vesicles (SUV) as a function of time of storage above their order-disorder phase transition. Vesicles composed of dipalmitoylphosphatidylcholine (DPPC) and dimyristoylphosphatidylcholine (DMPC) were examined. Cooling scans on fresh (5-7-h postsonication) samples revealed broad, relatively simple heat capacity peaks (peak temperatures: 19.9 degrees C for DMPC, 37.8 degrees C for DPPC) free of high-temperature spikes or shoulders. Subsequent heating scans displayed a sharp peak characteristic of previously described fusion products formed below the phase transition. SUV samples stored for 1 or more days above their phase transition displayed a moderately broad, high-temperature shoulder (23.8 degrees C for DMPC and 40.2 degrees C for DPPC) in the cooling profile. For DMPC, the enthalpy associated with this peak increased in a first-order fashion with time. Hydrolysis products were not detected until 12-20 days of storage. Both the rate and extent of shoulder appearance increased with temperature (k = 0.0017 h-1, fraction of total enthalpy = 0.1 at 36 degrees C; k = 0.0037 h-1, fraction = 0.2 at 42 degrees C). Freeze-fracture electron micrographs confirmed that an intermediate-sized vesicle population (diameters 400-500 A) appeared in SUV samples stored above their phase transition. Also, the trapped volume of DMPC SUV increased from 0.26 microL/mumol after 17 h of storage to 0.54 microL/mumol after storage for 16 days at 36 degrees C.(ABSTRACT TRUNCATED AT 250 WORDS)     

Mechanism of the spontaneous transfer of unconjugated bilirubin between small unilamellar phosphatidylcholine vesicles.

Unconjugated bilirubin (bilirubin-IX alpha), the hydrophobic end product of heme degradation, is esterified in the hepatocyte endoplasmic reticulum to water-soluble conjugates prior to excretion in bile. To characterize the process of intracellular bilirubin transport, the kinetic and thermodynamic activation parameters for the spontaneous transfer of bilirubin between small unilamellar egg lecithin vesicles were determined. Bilirubin-IX alpha was added to donor vesicles labeled with the fluorescent phospholipid probe, (5-(dimethylamino)naphthalene-1-sulfonyl) dipalmitoyl-L-alpha-phosphatidylethanolamine (dansyl-PE). When bound to the donor vesicles, bilirubin quenches the dansyl probe fluorescence through resonance energy transfer. The movement of bilirubin from dansyl-labeled donor vesicles to unlabeled acceptor vesicles was monitored directly by the reemergence of dansyl fluorescence over time. Vesicle fusion and intervesicle transfer of the dansyl-PE probe were excluded by quasielastic light scattering and fluorescence resonance energy transfer studies. Stopped-flow analysis demonstrated that the transfer of bilirubin was described by a single-exponential function with a mean half-time of 2.0 +/- 0.1 ms (+/- SD) at 37 degrees C. The rate of bilirubin transfer was independent of acceptor vesicle concentration and decreased with increasing buffer ionic strength, indicating that intermembrane transfer occurred via aqueous diffusion, rather than vesicle collisions. The free energy of activation (delta G++) for the dissociation of bilirubin from donor vesicles was 14.2 kcal.mol-1. These studies suggest that bilirubin is associated with phospholipid bilayers at the membrane-water interface. We postulate that the movement of unconjugated bilirubin between intracellular membranes occurs via spontaneous transfer through the aqueous phase.     

Micropolarities of lipid bilayers and micelles 5. Localization of pyrene in small unilamellar phosphatidylcholine vesicles

The excimer/monomer ratio of emission intensities (I_E/I_M) and the enhancement of the 0-0 vibronic transition in the fluorescence spectra of pyrene (PY) and 16-(1-pyrenyl)hexadecanoic acid (C₁₆PY) were used to investigate the localization of PY in the bilayers of small unilamellar vesicles constituted of phosphatidylcholine (SUV-PC). First, from comparison of the fluorescence characteristics of PY in water with those of PY incorporated into the SUV-PC membranes, we concluded that the probe is incorporated preferentially in the lipid phase of the vesicles and not in the bulk aqueous phase. In addition, we found that, contrary to what happens with the pyrenyl moiety of C₁₆PY the location of PY varies with its relative concentration in the membrane space. The critical concentration was observed to be around 1.0 mol% of incorporated PY. At concentrations below this value, PY is located in the hydrocarbon core of the lipid bilayers. Above 1.0 mol%, the PY molecules reside preferentially in the neighbourhood of the glyceryl moiety region of the PC vesicles.     

Location and dynamics of anthracyclines bound to unilamellar phosphatidylcholine vesicles

We have exploited the intrinsic fluorescence properties of the anthracycline antitumor antibiotics to study the dependence on drug structure of relative drug location and dynamics when the anthracyclines were bound to sonicated dimyristoylphosphatidylcholine (DMPC) and dipalmitoylphosphatidylcholine (DPPC) vesicles at 27.5 degrees C. Iodide quenching experiments at constant ionic strength were used to evaluate the relative accessibilities of the bound fluorophores to membrane-impermeable iodide. Iodide was found to quench the fluorescence of anthracyclines in free solution by both static and dynamic mechanisms, whereas quenching of membrane-bound fluorophores was predominantly due to the dynamic mechanism. Modified Stern-Volmer plots of anthracyclines bound to fluid-phase DMPC bilayers were linear, and the biomolecular rate constant (kq) values ranged from 0.6 X 10(9) to 1.3 X 10(9) M-1 s-1. Modified Stern-Volmer plots of anthracyclines bound to solid-phase DPPC bilayers were curved, indicative of a heterogeneous-bound drug population. A strong correlation between drug hydrophobicity and penetration of the fluorophore into the bilayer was observed for the daunosamine-containing anthracyclines. Steady-state fluorescence anisotropy measurements under iodide quenching conditions were used to investigate the diffusive motions of anthracyclines in isotropic solvent and in fluid-phase DMPC bilayers. Anthracycline derivatives free in solution exhibited limiting anisotropy (alpha infinity) values which decayed to zero at times long compared to the excited-state lifetime, in contrast to anthracyclines bound to fluid-phase DMPC bilayers, which showed nonzero alpha infinity values. Steady-state anisotropies of membrane-bound anthracyclines were found to be governed principally by alpha infinity and not by the mean rotational rate (R).(ABSTRACT TRUNCATED AT 250 WORDS)     

Exchangeability and rate of flip-flop of phosphatidylcholine in large unilamellar vesicles, cholate dialysis vesicles, and cytochrome oxidase vesicles.

Three model membrane systems have been characterized in terms of their interaction with phospholipid exchange proteins. Large unilamellar vesicles of phosphatidylcholine prepared by ether vaporization are shown to be homogeneous by gel filtration. Phospholipid exchange proteins from three sources are capable of catalyzing the rapid exchange of approximately half of the phospholipid from these vesicles. The remaining pool of radioactive phospholipid is virtually nonexchangeable (t1/2 of several days). Small unilamellar vesicles of phosphatidylcholine prepared by cholate dialysis also exhibit two pools of phospholipid (65% rapidly exchangable, 35% very slowly exchangeable) when incubated with beef liver phospholipid exchange protein. Cytochrome oxidase vesicles prepared both by a cholate dialysis method and by a direct incorporation method have been fractionated on a Ficoll discontinuous gradient, and tested for interaction with beef heart exchange protein. Two pools of phospholipid are once again observed (70% rapidly exchangable, 30% nonexchangeable), even for vesicles which have incorporated the transmembranous enzyme at a phospholipid to protein weight ratio of 2. The size of the rapidly exchangeable pool of phosphatidylcholine for each of the vesicle systems is consistent with the calculated fraction of phospholipid in the outer monolayer. The extremely slow rate of exchange of the second pool of the second pool of phospholipid reflects the virtual nonexistence of phospholipid flip-flop in any of these model membranes.     

A 13C NMR method for determination of the transbilayer distribution of phosphatidylcholine in large, unilamellar, protein-free and protein-containing vesicles.

(1) Large unilamellar vesicles have been prepared from N-[Ne3-13C]-18 : 1c/18 : 1c-phosphatidylcholine, both with and without the major intrinsic proteins from the human erythrocyte membrane incorporated in the bilayer. (2) It is shown that the inside-outside distribution of the lipid molecules in these large unilamellar structures can be determined using 13C NMR. (3) Large vesicles of 18 : 1c/18 : 1c-phosphatidylcholine containing glycophorin show an enhanced permeability to Dy3+. It is shown that the permeability barrier of these vesicles can be restored by addition of 10 mol% 18 : 1c/18 : 1c-phosphatidylethanolamine or 1-18 : 1c-lysophosphatidylcholine.     

Rapid transmembrane movement of phosphatidylcholine in small unilamellar lipid vesicles formed by detergent removal

Small unilamellar phosphatidylcholine vesicles, formed by solubilizing phosphatidylcholine with sodium cholate and removing the detergent by gel filtration, have been studied in their interaction with phospholipid exchange protein. The exchange of phosphatidylcholine between the vesicles and erythrocyte ghosts was greatly stimulated by the phosphatidylcholine-specific exchange protein from bovine liver. It was found that 95% of the phosphatidylcholine was readily available for exchange within 3 h at 37°C. In similar vesicles prepared by sonication only 70% of the phosphatidylcholine was rapidly exchangeable. Our results indicate that the transmembrane movement of phosphatidylcholine across the bilayer of vesicles prepared by the cholate technique is a relatively fast process. The results are discussed with respect to the presence of trace amounts of lipid-associated cholate which may facilitate the transbilayer exchange of phosphatidylcholine.     

Lysophosphatidylcholine stabilizes small unilamellar phosphatidylcholine vesicles. Phosphorus-31 NMR evidence for the "wedge" effect

Sonication of 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) and 1-palmitoyl-sn-glycero-3-phosphocholine (lysoPC, up to approximately 30 mol %) produces small unilamellar vesicles (SUV, 250-265 A diameter). Phosphorus-31 NMR of the POPC/lysoPC vesicles gives rise to four distinct peaks for POPC and lysoPC in the outer and in the inner bilayer leaflet which can be used to localize and quantify the phospholipids in both vesicle shells. Addition of paramagnetic ions (3 mM Pr3+) enhances outside/inside chemical shift differences and allows monitoring of membrane integrity by the absence of Pr3+ in the vesicle interior. 31P NMR shows that lysoPC in these highly curved POPC/lysoPC vesicles prefers the outer bilayer leaflet. LysoPC incorporation into POPC SUV furthermore causes a substantial and concentration-dependent decrease in spin-spin relaxations (T*2) of the outside POPC phosphorus signals from 55 ms for pure POPC vesicles (v1/2, 5.8 Hz) to 29.5 ms (v1/2, 10.8 Hz) for POPC/lysoPC vesicles containing 25 mol % lysoPC. Our findings are consistent with the idea of a cone-shaped lysoPC molecule which, for geometric reasons, is preferentially accommodated in the outer bilayer leaflet. LysoPC incorporation into POPC SUV restricts POPC headgroup motion and tightens phospholipid packing, but only in the outer bilayer shell.     

Transbilayer phosphatidylcholine distributions in small unilamellar sphingomyelin-phosphatidylcholine vesicles: effect of altered polar head group

The effect of the altered polar head group of phosphatidylcholine (PC) on its transbilayer distributions in small unilamellar vesicles containing sphingomyelin (SM) was ascertained with phospholipase A2 as the external membrane probe. These vesicles were formed by sonication and fractionated by centrifugation. The vesicle size was determined by gel-permeation chromatography and solute entrapment. Experiments were done to confirm that phospholipase A2 treatments did not induce fusion, lyse the vesicles, or cause PC to migrate across the vesicle bilayer. The complete degradation of external PC in intact vesicles was assured by carrying out the enzyme reactions in the absence as well as in the presence of 9.2 X 10(-5) M bovine serum albumin. In small vesicles comprised of SM and 30 mol % 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC), DPPC preferentially distributed in the inner monolayer. This preference of DPPC in these vesicles disappeared upon introducing one C2H5 group at the carbon atom adjacent to the quaternary ammonium residue in its polar head group and was reversed when the C2H5 group was replaced by C6H5 and C6H5CH2 substituents or when the P-N distance was increased. These results indicate that the effective polar head-group volume is an important factor in determining the phospholipid distributions across the small vesicle bilayer.     

On the size-dependent disintegration of small unilamellar phosphatidylcholine vesicles in rat plasma. Evidence of complete loss of vesicle structure.

The destruction of small unilamellar egg phosphatidylcholine vesicles in rat plasma was monitored by measuring release of encapsulated 125I-poly(vinylpyrrolidone) or carboxyfluorescein and by determining transfer of radiolabelled phosphatidylcholine to plasma lipoproteins by means of gel filtration. The susceptibility of the vesicles to the destructive action of plasma increased with decreasing vesicle size, as observed by incubating plasma with individual fractions constituting the small-vesicle peak on Sepharose CL-2B. This results in selective destruction of small vesicles when heterogeneous vesicle populations are incubated with plasma. Samples of homogeneous vesicle populations were incubated with a wide range of plasma concentrations, which resulted in extents of solute and phospholipid release ranging from 10 to 90%. When the extents of solute release were plotted against the extents of lipid release a linear, virtually 1:1, relationship was found, for both carboxyfluorescein and poly(vinylpyrrolidone) as the solute. This suggests that the release of solutes from small unilamellar phosphatidylcholine vesicles as a result of their interaction with plasma (lipo)proteins involves the total destruction of a fraction of the vesicles, the magnitude of which is determined by the vesicle: plasma ratio. Our results argue against a previously presented view suggesting that the interaction between such vesicles and plasma results in the formation of pores through which encapsulated solutes diffuse at Mr-dependent rates [Kirby & Gregoriadis (1981) Biochem. J. 199, 251-254]. The discrepancies between the two studies in observations as well as in interpretation are discussed.     

Assembly of the membrane attack complex of complement on small unilamellar phospholipid vesicles

Light-scattering intensity was shown to be a reliable, direct, and quantitative technique for monitoring the assembly of the membrane attack complex of complement (proteins C5b-6, C7, C8, and C9) on small unilamellar phosphatidylcholine vesicles. The assembly on vesicles occurred in a simple fashion; complexes of C5b-7 bound noncooperatively to the vesicles, and final assembly of C5b-9 did not induce vesicle aggregation or fragmentation. When C5b-6 and C7 were mixed in the presence of vesicles but at molar protein/vesicle ratios of less than 1, there was quantitative binding of C5b-7 to the vesicles with no concomitant aggregation of C5b-7. If C7 was added at a slower rate, quantitative binding was obtained at molar C5b-7/vesicle ratios of up to 5. The latter observations (a) were consistent with the proposal that C5b-7 aggregation and membrane binding were competitive events and (b) defined conditions under which light-scattering intensity measurements could monitor C5b-9 assembly on vesicles without contribution from the fluid-phase assembly. The C8/C5b-7 ratio in the phospholipid-C5b-8 complex was 0.97 +/- 0.12, and the maximum ratio of C9/C5b-8 in the final complex was 16.2 +/- 2.0. One C9 molecule associated rapidly with each phospholipid-C5b-8, followed by slower incorporation of the remaining C9 molecules. The initial velocity of the slow phase of C9 addition was easily saturated with C9 and gave an activation energy of 37 kcal/mol. This was identical with the value measured for the analogous process in the fluid-phase assembly.(ABSTRACT TRUNCATED AT 250 WORDS)     

A simple method for the preparation of small unilamellar vesicles

A simple and quick method for the preparation of small unilamellar vesicles (SUV) was developed. SUV are spontaneously formed by swelling of the specially prepared phospholipid film in water/buffer. Normally, large multilamellar vesicles (MLV) are formed when a phospholipid film is dissolved in water. To prevent the formation of multilamellar structures we used the slightly charged phospholipids which exhibit infinite swelling while the formation of large structures was prevented by the deposition of the phospholipid film on the support with small surfaces. These two requirements were met by mixing a small amount of ionic detergent into phospholipid which was deposited on microcrystals. The size and size distribution of the produced vesicles depend on the size and homogeneity of the microcrystals. When 1.5 wt% of cetyltetramethylammonium bromide (CTAB) in egg yolk phosphatidylcholine was deposited on zeolite X microcrystals with crystallite sizes of approx. 0.4 μm a homogeneous population of vesicles with average diameter 21.5 nm was obtained.     

Binding of the antibacterial peptide magainin 2 amide to small and large unilamellar vesicles

The thermodynamics of binding of the antibacterial peptide magainin 2 amide (M2a) to negatively charged small (SUVs) and large (LUVs) unilamellar vesicles has been studied with isothermal titration calorimetry (ITC) and CD spectroscopy at 45 degrees C. The binding isotherms as well as the ability of the peptide to permeabilize membranes were found to be qualitatively and quantitatively similar for both model membranes. The binding isotherms could be described with a surface partition equilibrium where the surface concentration of the peptide immediately above the plane of binding was calculated with the Gouy-Chapman theory. The standard free energy of binding was deltaG0 approximately -22 kJ/mol and was almost identical for LUVs and SUVs. However, the standard enthalpy and entropy of binding were distinctly higher for LUVs (deltaH0 = -15.1 kJ/mol, deltaS0 = 24.7 J/molK) than for SUVs (deltaH0 = -38.5 kJ/mol, deltaS0 = -55.3 J/molK). This enthalpy-entropy compensation mechanism is explained by differences in the lipid packing. The cohesive forces between lipid molecules are larger in well-packed LUVs and incorporation of M2a leads to a stronger disruption of cohesive forces and to a larger increase in the lipid flexibility than peptide incorporation into the more disordered SUVs. At 45 degrees C the peptide easily translocates from the outer to the inner monolayer as judged from the simulation of the ITC curves.     

Ca2+-induced fusion of sulfatide-containing phosphatidylethanolamine small unilamellar vesicles.

The fusogenic properties of sulfatide-containing 1,2-dioleoyl-3-sn -phosphatidylethanolamine (DOPE) small unilamellar vesicles (SUVs) in the presence of CaCl2 were studied by mixing membrane lipids based on an assay of fluorescence resonance energy transfer (FRET). Fusion of the vesicles was also confirmed by mixing aqueous contents with the Tb/dipicolinate (DPA) assay. The half-times of lipid mixing revealed that the fusion rate decreased with increasing molar concentration of sulfatide. This inhibitory effect was more obvious at sulfatide concentrations higher than 30 mol%, where hydration at the membrane surface reached its maximum and the fusion was no longer pH-sensitive in the range of pH 6.0 - 9.0. Similar inhibitory effect was also observed in Ca2+-induced fusion of DOPE/ganglioside GM1 vesicles but at a lower concentration of the glycosphingolipid (20 mol%). In contrast, increasing the concentration of phosphatidylserine (PS) in DOPE/PS SUVs resulted in an increase in the rate of Ca2+-induced lipid mixing and the pH sensitivity of this system was not affected.These results are consistent with an increasing steric hindrance to membrane fusion at higher molar concentration and larger headgroup size of the glycosphingolipids. Interestingly, the pH sensitivity of the sulfatide-containing liposomes was retained when they were allowed to fuse with synaptosomes in the absence of Ca2+ by a mechanism involving protein mediation.     

Hydrolysis of dipalmitoylphosphatidylcholine small unilamellar vesicles by porcine pancreatic phospholipase A2

The hydrolysis of small unilamellar vesicles made of dipalmitoylphosphatidylcoline by pancreatic phospholipase A2 has been studied under various conditions of temperature and enzyme and substrate concentration using the following three different experimental protocols. When the enzyme was added to the substrate vesicles after being separately adjusted to the temperature of the experiments hydrolysis occurred instantaneously only in the temperature range where the lipid is known to exist in its gel phase, while above the transition range no hydrolysis occurred. Within the transition range, the time course of hydrolysis was characterized by initial very slow rate of hydrolysis (latency phase) followed by an abrupt increase in the rate after a time tau, which is a complex function of temperature and enzyme to substrate ratio. When an enzyme-substrate mixture was first preincubated below Tm and then temperature jumped to a temperature above or within the transition range, the latency phase was markedly shortened. When the temperature jump was to the transition range, this effect is observed even if Ca2+ is absent in the preincubation mixture. However, instantaneous hydrolysis was observed upon temperature jumping the mixture to a temperature high above Tm only if Ca2+ was present in the preincubation medium. In temperature-scanning experiments, hydrolysis was followed while changing the temperature of the enzyme-substrate mixture continuously. Heating an enzyme-substrate mixture from room temperature resulted in an abrupt onset of hydrolysis when the transition range was approached. These results lead us to conclude that two distinctly different steps precede rapid hydrolysis of dipalmitoylphosphatidylcholine small unilamellar vesicles by pancreatic phospholipase A2: a Ca2+-independent binding of the enzyme to the substrate vesicles, which for chemically pure bilayers occurs best in the gel phase. This step is followed by a Ca2+-dependent activation of the initially formed enzyme-substrate complex. The latter step only occurs under conditions where the bilayer possesses packing irregularities and probably involves a reorganization of the enzyme-substrate complex. At least one of these two steps appears to involve enzyme-enzyme interaction.     

Stability of giant unilamellar vesicles and large unilamellar vesicles of liquid-ordered phase membranes in the presence of Triton X-100

We have investigated the stability of giant unilamellar vesicles (GUVs) and large unilamellar vesicles (LUVs) of lipid membranes in the liquid-ordered phase (lo phase) against a detergent, Triton X-100. We found that in the presence of high concentrations of Triton X-100, the structure of GUVs and LUVs of dipalmitoyl-PC (DPPC)/cholesterol (chol) and sphingomyelin (SM)/chol membranes in the lo phase was stable and no leakage of fluorescent probes from the vesicles occurred. We also found that ether-linked dihexadecylphosphatidylcholine (DHPC) membranes containing more than 20 mol% cholesterol were in the lo phase, and that DHPC/chol-GUV and DHPC/chol-LUV in the lo phase were stable and no leakage of internal contents occurred in the presence of Triton X-100. In contrast, octylglucoside solution could easily break these GUVs and LUVs of the lo phase membranes and induced internal contents leakage. These data indicate that GUVs and LUVs of the lo phase membranes are very valuable for practical use.     

pH-dependent membrane fusion and vesiculation of phospholipid large unilamellar vesicles induced by amphiphilic anionic and cationic peptides.

We studied fusion induced by a 20-amino acid peptide derived from the amino-terminal segment of hemagglutinin of influenza virus A/PR/8/34 [Murata, M., Sugahara, Y., Takahashi, S., & Ohnishi, S. (1987) J. Biochem. (Tokyo) 102, 957-962]. To extend the study, we have prepared several water-soluble amphiphilic peptides derived from the HA peptide; the anionic peptides D4, E5, and E5L contain four and five acidic residues and the cationic peptide K5 has five Lys residues in place of the five Glu residues in E5. Fusion of egg phosphatidylcholine large unilamellar vesicles induced by these peptides is assayed by two different fluorescence methods, lipid mixing and internal content mixing. Fusion is rapid in the initial stage (12-15% within 20 s) and remains nearly the same or slightly increasing afterward. The anionic peptides cause fusion at acidic pH lower than 6.0-6.5, and the cationic peptide causes fusion at alkaline pH higher than 9.0. Leakage and vesiculation of vesicles are also measured. These peptides are bound and associated with vesicles as shown by Ficoll discontinuous gradients and by the blue shift of tryptophan fluorescence. They take an alpha-helical structure in the presence of vesicles. They become more hydrophobic in the pH regions for fusion. When the suspension is made acidic or alkaline, the vesicles aggregate, as shown by the increase in light scattering. The fusion mechanism suggests that the amphiphilic peptides become more hydrophobic by neutralization due to protonation of the carboxyl groups or deprotonation of the lysyl amino groups, aggregate the vesicles together, and interact strongly with lipid bilayers to cause fusion. At higher peptide concentrations, E5 and E5L cause fusion transiently at acidic pH followed by vesiculation.