Phospholipid Composition and Metabolism of Micrococcus denitrificans

The phospholipid composition of Micrococcus denitrificans was unusual in that phosphatidyl choline (PC) was a major phospholipid (30.9%). Other phospholipids were phosphatidyl glycerol (PG, 52.4%), phosphatidyl ethanolamine (PE, 5.8%), an unknown phospholipid (5.3%), cardiolipin (CL, 3.2%), phosphatidyl dimethylethanolamine (PDME, 0.9%), phosphatidyl monomethylethanolamine (PMME, 0.6%), phosphatidyl serine (PS, 0.5%), and phosphatidic acid (0.4%). Kinetics of ³²P incorporation suggested that PC was formed by the successive methylations of PE. Pulse-chase experiments with pulses of ³²P or acetate-1-¹⁴C to exponentially growing cells showed loss of isotopes from PMME, PDME, PS, and CL with biphasic kinetics suggesting the same type of multiple pools of these lipids as proposed in other bacteria. The major phospholipids, PC, PG, and PE, were metabolically stable under these conditions. The fatty acids isolated from the complex lipids were also unusual in being a simple mixture of seven fatty acids with oleic acid representing 86% of the total. Few free fatty acids and no non-extractable fatty acids associated with the cell wall or membrane were found.     

The influence of ionic detergents on the phospholipid fatty acid compositions of Porphyridium purpureum

The phospholipid fatty acid composition of Porphyridium purpureum on a solid medium was studied in the presence of sodium dodecyl sulphate (SDS) and cetyl trimethylammonium bromide (CTAB). The most common fatty acids in phosphatidyl choline (PC) and phosphatidyl ethanolamine (PE) were palmitic (16:0), stearic (18: 0), linoleic (18:2ω 6), arachidonic (20:4ω 6) and eicosapentaenoic (20:5ω 3) acids, 20:4ω 6 being very abundant. In phosphatidyl glycerol (PG) the most common acids were 16:0, trans-hexadecenoic acid (tr 16:1ω 13), oleic acid (18:1) and 20:4ω 6. Both detergents increased the saturation grade of PC and PE by decreasing the relative amount of the polyunsaturated acids, especially 20:4ω 6. A corresponding increase in the amounts of saturated acids was observed in PC and PE. The changes in PG fatty acid composition were not very significant: a slight increase was observed in the amounts of 16:0 and tr 16:1ω 13 , with a corresponding decrease in the amounts of 20:4ω 6 and 20:5ω 3. Both detergents decreased the PC/PE and the (PC + PE)/PG ratios very markedly, most probably as a result of increases in the amounts of PE and PG. In the presence of CTAB the cells seemed to contain much more phospholipids than in the presence of SDS, perhaps as a result of the mucilage-precipitating effect of CTAB. The significance of the findings is discussed.     

Interactions between phospholipids and NADH:ubiquinone oxidoreductase (complex I) from bovine mitochondria

NADH:ubiquinone oxidoreductase (complex I) from bovine heart mitochondria is a highly complicated, energy transducing, membrane-bound enzyme. It contains 46 different subunits and nine redox cofactors: a noncovalently bound flavin mononucleotide and eight iron-sulfur clusters. The mechanism of complex I is not known. Mechanistic studies using the bovine enzyme, a model for human complex I, have been precluded by the difficulty of preparing complex I which is pure, monodisperse, and fully catalytically active. Here, we describe and characterize a preparation of bovine complex I which fulfills all of these criteria. The catalytic activity is strongly dependent on the phospholipid content of the preparation, and three classes of phospholipid interactions with complex I have been identified. First, complex I contains tightly bound cardiolipin. Cardiolipin may be required for the structural integrity of the complex or play a functional role. Second, the catalytic activity is determined by the amounts of phosphatidylcholine (PC) and phosphatidylethanolamine (PE) which are bound to the complex. They are more weakly bound than cardiolipin, exchange with PC and PE in solution, and can substitute for one another. However, their nontransitory loss leads to irreversible functional impairment. Third, phospholipids are also required in the assay buffer for the purified enzyme to exhibit its full activity. It is likely that they are required for solubilization and presentation of the hydrophobic ubiquinone substrate.     

Further study on the role of Mg2+ in lipid-protein interaction in reconstituted porcine heart mitochondrial H+-ATPase

Porcine heart mitochondrial H+-ATPase was reconstituted by cholate dialysis method in liposomes containing neutral (PC, PE), acidic (PG, PI, PA, PS, DPG) or neutral and acidic phospholipids. The Mg2+ effect on the ATPase activity and its sensitivity to oligomycin, ATP-induced delta psi and delta pH formation was observed for the proteoliposomes containing acidic but not neutral phospholipids. Maleimide spin labels with varying arm lengths or bromoacetamide spin probe were used to monitor the conformational difference of H+-ATPase in the Mg2+-containing and Mg2+-'free' samples. A difference in W/S ratio (weakly immobilized/strongly immobilized component in the ESR spectra) could be detected for the F0.F1-containing and F1-depleted, (F0)-containing proteoliposomes, suggesting conformational difference in the F0-F1 complex and F0 portion induced by the Mg2+ effect. A conformational change of the beta-subunits in the F1 portion was also deduced from the ATP-induced fluorescence quenching of aurovertin-complex for Mg2+-containing samples. The results obtained are in favor of our previous assumption that Mg2+ may play its role by altering the physical state of the lipid bilayer, which would induce a conformational change in F0 (buried in the lipid core), which in turn is transmitted to the catalytic F1, resulting in a higher enzyme activity.     

Cardiolipin-Specific Phospholipase D Activity in Haemophilus parainfluenzae

A highly active phospholipase D that is specific for cardiolipin was detected in the gram-negative bacterium Haemophilus parainfluenzae. Previously reported phospholipase D preparations have come exclusively from higher plants. The bacterial enzyme hydrolyzed cardiolipin to phosphatidyl glycerol and phosphatidic acid. During the incubation, phosphatidic acid disappeared. Phosphatidyl ethanolamine, methylated phosphatidyl ethanolamines, phosphatidyl choline, and phosphatidyl glycerol were not hydrolyzed when cardiolipin was rapidly hydrolyzed.     

Calcium-induced phase separation phenomena in multicomponent unsaturated lipid mixtures

The ability of calcium to induce phase separation in multicomponent lipid mixtures containing various unsaturated species of acidic and neutral phospholipids has been investigated by 31P NMR, 3H NMR, and small-angle X-ray diffraction techniques. It is shown that, in unsaturated (dioleoyl-) phosphatidylglycerol (PG)/phosphatidylethanolamine (PE) (1:1) and phosphatidic acid (PA)/phosphatidylcholine (PC) (1:1) mixtures, calcium is unable to induce lateral phase separation of the acidic and neutral lipids and that all the lipids adopt a hexagonal (HII) phase in the presence of calcium. In multicomponent mixtures containing one or more acidic species the presence of cholesterol either facilitates calcium-induced lamellar to hexagonal (HII) transitions for all the lipid components or, in systems already in a hexagonal (HII) phase, mitigates against calcium-induced lateral phase separations. Further, cholesterol is shown to exhibit no preferential interaction on the NMR time scale with either PC, PE, or phosphatidylserine (PS) when the lipids are in the liquid-crystal state. The ability of cholesterol to directly induce HII phase formation in PC/PE mixtures is also shown to be common to various other sterols including ergosterol, stigmasterol, coprostanol, epicoprostanol, and androstanol.     

Phospholipid dependence of rat brain microsomal glucose-6-phosphate phosphohydrolase

Partial lipid removal of rat brain microsomes by acetone-butanol extraction resulted in 32% loss of activity of glucose-6-phosphate phosphohydrolase (G-6-Pase) and an increase in Km and energy of activation (Ea) of the enzyme while the Vmax was lowered. The activity was restored by supplementation of microsomal total phospholipid (PL) and phosphatidylcholine (PC) in sonicated dispersions but not with neutral lipids, phosphatidyl ethanolamine, sphingomyelin, phosphatidylglycerol and cholesterol. In both intact and delipidated membranes, the activity was decreased by sodium deoxycholate and enhanced by dimethylsulfoxide. Egg yolk PC and asolectin influenced the activity to the extent of that produced by microsomal PC. PC increased the Km of the enzymatic reaction in intact microsomes but decreased the same in disrupted membrane while the Vmax was not affected in both the membranes. Addition of PC into the assay system lowered Ea of the reaction in both the membrane systems. However, there was no break observed in the Arrhenius plot. Ability of liver nonspecific lipid transfer proteins to introduce alien PL into brain microsomes was used to study lipid dependence of G-6-Pase and investigation of membrane-enzyme interrelationship. Protein catalyzed transfer of egg PC from a donor PC-cholesterol unilamellar liposomes resulted in substantial increase in microsomal membrane PC and total PL and a net reduction in the enzyme activity was observed in intact and delipidated membranes.(ABSTRACT TRUNCATED AT 250 WORDS)     

Effect of phospholipids on the activity of sialosyl lactosylceramide (GM3): N-acetylgalactosaminyl transferase from chick embryo brain

Summary A preparation of UDP-Ga1NAc:sialosyl lactosylceramide N-acetylgalactosaminyl transferase (E.C. 2.4.1.92) obtained from 14-day chick embryo brain was delipidated partially by treatment with cold acetone in the presence of varying amounts of sodium dodecyl sulphate (SDS). The lipid content and the enzyme activity of the preparation decreased as the concentration of SDS increased. At 0.3% SDS the lipid content was about 30% and the enzyme activity about 15% of the original. The activity could be restored up to 60% of the original by added phospholipids, provided the removal of endogenous lipids did not exceed 70%. Phospholipids with diferent composition showed different abilities to restore the enzyme activity. Among phosphatidylcholines the decreasing order of effectivity was dilauroyl dimiristoyl dipalmitoyl distearoyl-choline. Dimiristoyl phosphatidylcholine, dimiristoyl phosphatidylglycerol and dipalmitoyl phosphatidylglycerol activated the enzyme more effectively than dimiristoyl phosphatidyl ethanolamine, dimiristoyl phosphatidic acid or brain phosphatidylserine. No correlation was found between the activating ability and the charge in the polar head group of the lipid added. Addition of dilauroyl phosphatidylcholine to the delipidated preparation increased about 5-fold the Vmax without affecting the apparent Km for both the donor nucleotide and acceptor glycolipid. The data suggest that the lipid composition of the enzyme environment constitutes a potential level of regulation of the activity of this key enzyme of ganglioside biosynthesis.     

Phospholipase A2 from rat spleen lymphocytes hydrolyzing arachidonoyl-phospholipids]

Phospholipase A2 activity in the postnuclear supernatant of lymphocytes has been studied by measuring 14C arachidonate released from labelled phosphatidyl ethanolamine (PE) and phosphatidyl choline (PC) as exogenous substrates. The pH optimum was 7.5-9.0 for PE and 9.0 for PC. Phospholipase A2 was not detected in the presence of 2 mM EGTA. It was optimal with the millimolar calcium concentrations and higher towards PE. Preincubation of lymphocytes with 0.5 M ionophore A-23187 was followed by 2.4 fold stimulation of the phospholipase activity. A stimulatory effect was observed after preincubation of cells with 10 micrograms/ml of phytohemagglutinin, lipopolysaccharide, concanavalin A; it decreased as: lipopolysaccharide greater than phytohemagglutinin greater than concanavalin A. The results obtained have suggested the possibility of existence of different forms of phospholipase A2 in the spleen lymphocytes and participation of the enzyme in the early signalling events.     

Oxidoreductase activity of chromatophores and purified cytochrome bc 1 complex from Rhodobacter sphaeroides: a possible role of cardiolipin

Osmotic shock was used as a tool to obtain cardiolipin (CL) enriched chromatophores of Rhodobacter sphaeroides. After incubation of cells in iso- and hyper-osmotic buffers both chromatophores with a physiological lipid profile (Control) and with an almost doubled amount of CL (CL enriched) were isolated. Spectroscopic properties, reaction centre (RC) and reducible cytochrome (cyt) contents in Control and CL enriched chromatophores were the same. The oxidoreductase activity was found higher for CL enriched than for Control chromatophores, raising from 60?±?2 to 93?±?3?mol cyt c s(-1) (mol total cyt c)(-1). Antymicin and myxothiazol were tested to prove that oxidoreductase activity thus measured was mainly attributable to the cyt bc ( 1 ) complex. The enzyme was then purified from BH6 strain yielding a partially delipidated and almost inactive cyt bc ( 1 ) complex, although the protein was found to maintain its structural integrity in terms of subunit composition. The ability of CL in restoring the activity of the partially delipidated cyt bc ( 1 ) complex was proved in micellar systems by addition of exogenous CL. Results here reported indicate that CL affects oxidoreductase activity in the bacterium Rhodobacter sphaeroides both in chromatophore and in purified cyt bc ( 1 ) complex.     

The effect of adriamycin and duramycin on calcium translocation in liposome systems modeled on the inner mitochondrial membrane

Adriamycin (doxorubicin, AdM) is a potent antineoplastic agent which binds specifically and with high affinity to the acidic phospholipid cardiolipin (CL) [Goormaghtigh et al. (1980) Biochim. Biophys. Acta 597, 1]. Duramycin (DM), a polypeptide antibiotic, has been reported to interact selectively with phosphatidylethanolamine (PE) and monogalactosyldiacylglycerol [Navarro et al. (1985) Biochemistry 24, 4645]. The selectivity of DM-PE interaction was confirmed. AdM and DM were then used to explore the roles of CL and PE in Ca2+ translocation in a phosphatidylcholine (PC)/PE/CL liposome system modeled on the inner mitochondrial membrane with the following results: (i) AdM (100-400 microM) altered Ca2+ uptake by PC/PE/CL (4/4/1, mol/mol) liposomes in a concentration-dependent fashion which varied with temperature, external Ca2+ concentration, and liposome PE content. (ii) Addition of AdM was qualitatively equivalent to increasing temperature, Ca2+ concentration, or liposome PE content, and cooperative interactions among these parameters were observed. An increase in any one factor generally enhanced Ca2+ uptake; simultaneous increases in several factors inhibited uptake. (iii) Inhibition of Ca2+ uptake was correlated with efflux of Arsenazo III. (iv) Ca2+ uptake by PC/PE/CL liposomes is biphasic [Kester and Sokolove (1989) Biochim. Biophys. Acta 980, 127]. DM suppressed the PE-dependent slow phase and stimulated the PE-independent initial phase. Ca2+ uptake by PC/PE/CL liposomes in the presence of DM resembled uptake by PC/CL liposomes. These data confirm the ability of PE to enhance the slow, highly temperature-dependent component of CL-mediated Ca2+ translocation and suggest that this process is sensitive to lipid phase behavior.     

Nonspecific phospholipase C of Listeria monocytogenes: activity on phospholipids in Triton X-100-mixed micelles and in biological membranes.

Listeria monocytogenes secretes a phospholipase C (PLC) which has 39% amino acid sequence identity with the broad-specificity PLC from Bacillus cereus. Recent work indicates that the L. monocytogenes enzyme plays a role during infections of mammalian cells (J.-A. Vazquez-Boland, C. Kocks, S. Dramsi, H. Ohayon, C. Geoffroy, J. Mengaud, and P. Cossart, Infect. Immun. 60:219-230, 1992). The homogeneous enzyme has a specific activity of 230 mumol/min/mg when phosphatidylcholine (PC) is dispersed in sodium deoxycholate. With phospholipid-Triton X-100 mixed micelles, the enzyme had a broad pH optimum between 5.5 and 8.0, and the rates of lipid hydrolysis were in the following order: PC > phosphatidylethanolamine (PE) > phosphatidylserine > sphingomyelin >> phosphatidylinositol (PI). Activity on PC was stimulated 35% by 0.5 M NaCl and 60% by 0.05 mM ZnSO4. When Escherichia coli phospholipids were dispersed in Triton X-100, PE and phosphatidylglycerol, but not cardiolipin, were hydrolyzed. The enzyme was active on all phospholipids of vesiculated human erythrocytes including PI, which was rapidly hydrolyzed at pH 7.0. PI was also hydrolyzed in PI-PC-cholesterol liposomes by the nonspecific PLC from L. monocytogenes and by the homologous enzyme from B. cereus. The water-soluble hydrolysis product was identified as inositol-1-phosphate. For the hydrolysis of human erythrocyte ghost phospholipids, a broad pH optimum was also observed. 32P-labelled Clostridium butyricum protoplasts, which are rich in ether lipids, were treated with PLC. The enzyme hydrolyzed the plasmalogen form of PE, its glycerol acetal, and cardiolipin, in addition to PE. I-, Cl- and F- stimulated activity on either PC- Triton X-100 mixed micelles or human erythrocyte ghosts, unlike the enzyme from B. cereus which is strongly inhibited by halides. Tris-HCl, phosphate, and calcium nitrate had similar inhibitory effects on the enzyme on the enzymes from L. monocytogenes and B. cereus.     

Biotinylated Stealth magnetoliposomes

Dimyristoylphosphatidylethanolamine (DC(14:0)PE) and the dioleoyl analogue (DC(18:1cis)PE) were mixed with alpha-biotinylamido-omega-N-succinimidoxycarbonyl-poly(ethylene glycol) (NHS-PEG-biotin) and quantitatively converted to alpha-biotinylamido-omega-(dimyristoylphosphatidylethanolamino-carbonyl)polyethylene glycol (DC(14:0)PE-PEG-biotin) and the dioleoyl analogue DC(18:1cis)PE-PEG-biotin, respectively. As shown by thin-layer chromatography and 1H NMR spectroscopy, PEGylation of both phosphatidylethanolamine types went to completion if the reaction was performed in organic solvent in the presence of triethylamine. The resulting derivatives were successfully incorporated into both classical phospholipid vesicles and a phospholipid bilayer surrounding nanometer-sized magnetite cores. In the latter case, the so-called activated Stealth(1) magnetoliposomes were produced which very efficiently immobilized streptavidinylated alkaline phosphatase.     

Phospholipase-mediated preparation of 1-ricinoleoyl-2-acyl-sn-glycero-3-phosphocholine from soya and egg phosphatidylcholine

1-Ricinoleoyl-2-acyl-sn-glycero-3-phosphocholine was prepared by incorporating ricinoleic acid completely in the sn-1 position of egg and soya phosphatidylcholine (PC) using immobilized phospholipase A(1) as the catalyst. The optimum reaction conditions for maximum incorporation of ricinoleic acid into PC through transesterification were 10% (w/w) immobilized enzyme (116 mg), a 1:5 mol ratio of PC (soya, 387 mg; egg, 384 mg) to methyl ricinoleate (780 mg) at 50 degrees C for 24 h in hexane.     

Control of the potassium ion-stimulated adenosine triphosphatase of pig gastric microsomes: effects of lipid environment and the endogenous activator

The K⁺-stimulated ATPase activity associated with the purified gastric microsomes from the pig gastric mucosa can be completely inactivated by treatment with 15% ethanol for 60 s at 37 °C but not at 25 °C. Sequential exposure of the microsomes to 15% ethanol at 25 and 37 °C caused the release of 2.9 and 4.3% of the total membrane phospholipids, respectively, consisting entirely of phosphatidyl choline and phosphatidyl ethanolamine. The ethanol-treated (37 °C) membrane had high basal (with Mg²⁺ as the only cation in the assay mixture) activity, which was further enhanced during reconstitution with phosphatidyl choline or phosphatidyl ethanolamine. The high basal activities could be reduced to the normal control level by assaying the enzyme in presence of the “activator protein,” partially purified from the soluble supernatant of the pig gastric cells. Phosphatidyl choline was somewhat more effective than phosphatidyl ethanolamine in the restoration of the activity of the ethanol-treated enzyme while phosphatidyl serine, phosphatidyl inositol, and sphingomyelin were without any effect. Synthetic phosphatidyl choline with various fatty acid substitutions were tested for their effectiveness in the restoration of the ethanol-inactivated enzyme. The distearoyl (18:0), dioleoyl (18:1), and dilinoleoyl (18:2) derivatives of phosphatidyl choline were almost equally effective while dipalmitoyl (16:0) phosphatidyl choline was somewhat less effective in the reconstitution process. Cholesterol appeared to interfere with phosphatidyl choline in the restoration of the activity of ethanol-treated enzyme. The fatty acid composition of phosphatidyl choline and phosphatidyl ethanolamine extracted by 15% ethanol at 37 °C was clearly different than those of the total microsome. Our data suggest that the phospholipids extracted by 15% ethanol at 37 °C are derived primarily from the immediate lipid environment of the enzyme and ATP together with Mg²⁺ and K⁺ help the partially delipidated enzyme to retain the appropriate conformation for the subsequent reconstitution. Furthermore, ethanol appears to either release or inactivate the membrane-associated activator protein, demonstrated to be essential for the K⁺-stimulated activity of the pig gastric ATPase.     

Gibberellin modulation of phosphatidyl-choline turnover in wheat aleurone tissue

Phosphatidyl choline (PC) is synthesised in wheat (Triticum aestivum L. cv. Flanders) aleurone tissue during early germination when new endomembranes are being formed. Although gibberellic acid does not ostensibly affect PC levels, it inhibits the incorporation of choline and differentially and specifically modulates the turnover of the N-methyl and methylene carbons of the choline headgroup of PC. Gibberellic acid has no effect on turnover of the phosphate moiety of either PC or the other major phosphatides. The possible biological importance of the findings is discussed.Abbreviations ER endoplasmic reticulum - GA gibberellin - GA₃ gibberellic acid - PA phosphatidic acid - PC phosphatidyl choline - PE phosphatidyl ethanolamine - PG phosphatidyl glycerol - PI phosphatidyl inositol - t_1/2 half-life     

31P MRS analysis of the phospholipid composition of normal human peripheral blood mononuclear cells (PBMC)

The aim of this investigation was to characterize the phospholipid composition of normal human blood mononuclear cells using 31P NMR spectroscopy. Mononuclear cells of peripheral blood were obtained from 10 volunteers. Phospholipid extracts were prepared from 60x10(6) cells according to modified Folch's method. An AMX 300 Bruker spectrometer 7.05 T was used. The 31P spectrum of phospholipid extracts from normal human PBMC consisted of 9 peaks, with one each for phosphatidylcholine (PC), plasmalogen of phosphatidylcholine (CPLAS), lysophosphatidylcholine (LPC), sphingomyelin (SM), phosphatidylethanolamine (PE), phosphatidylinositol (PI), phosphatidylserine (PS) and cardiolipin (CL), and another one due to the external reference substance, methylenediphosphonic acid (MDPA). The concentrations of these phospholipids (PL), based on the integral intensities, were as follows: 0.398 +/- 0.078 mmole/l for PC; 0.033 +/- 0.019 mmole/l for CPLAS; 0.155 +/- 0.043 mmole/l for SM; 0.266 +/- 0.104 mmole/l for PI+PE; 0.101 +/- 0.040 mmole/l for PS, and 0.026 +/- 0.033 mmole/l for CL. The results of this study confirmed that 31P MRS is a convenient tool for measuring the phospholipid concentrations of biological samples.     

Changes in phospholipids and free fatty acids in the brains of mice preconditioned by hypoxia.

Changes in the composition and contents of phospholipids and free fatty acids were observed and compared in three groups: (A) unpreconditoned normal controls, (B) exposure to 1 run of hypoxia and (C) exposure to 4 runs of hypoxia. In group B, the content of phosphatidyl ethanolamine (PE), phosphatidyl serine (PS) and free fatty acids (FFAs) increased significantly and the content of phosphatidyl choline (PC) and sphingomyelin (SM) decreased significantly. While in group C the content of PE, PS, PC and FFAs changed significantly when compared with that of group B, all phospholipid (except SM) and FFA contents tended to decrease to the level of group A. No new FFA was seen in the brain homogenates in any of the three groups. These results suggest that the changes in the content of mouse brain phospholipids and FFAs may be adaptive and involved in the animals' tolerance to hypoxia.     

Interaction of the water-soluble protein aprotinin with liposomes: gel-filtration, turbidity studies, and 31P NMR studies

The interactions of a water-soluble nonmembrane protein aprotinin with multilamellar vesicles (MLV) and small unilamellar vesicles (SUV) from soybean phospholipids were studied using Sephadex G-75 gel chromatography combined with different methods of the analysis of the eluate fractions (fluorescence, light-scattering, turbidity; 31P NMR spectroscopy). The composition of the liposomes mainly containing soybean phosphatidylcholine (PC) was varied by the addition of phosphatidylethanolamine (PE), phosphatidylinositol (PI) and lyso-phosphatidylcholine (lyso-PC). To evaluate the lipid-protein interactions, the amount of aprotinin in the MLV-aprotinin complexes was determined. Lipid-protein interactions were found to strongly depend on the liposome composition, medium pH and ionic strength. These dependencies point to the electrostatic nature of the aprotinin-lipid interactions. 31P NMR spectroscopy of the MLV-aprotinin complexes indicated that aprotinin influences the phospholipid structure in MLV at pH 3.0. In the case of PC:PE:PI and PC:PE:PI:lyso-PC vesicles, aprotinin induced liposome aggregation and a lamellar-to-isotropic phase transition of the phospholipids.