Geometric packing constraints in egg phosphatidylcholine vesicles

Small phospholipid vesicles: internal pressure, surface tension, and surface free energy

Tanford [Tanford, C. (1979) Proc. Natl. Acad. Sci. USA 76, 3318-3319] used thermodynamic arguments to show that the pressure difference across the bilayer of small phospholipid vesicles must be zero. This paper analyzes the implications of this conclusion in terms of Laplace's law and the basic thermodynamics of interfaces. In its usual form, Laplace's law is of questionable value for the vesicle. If the vesicle is in a state of metastable equilibrium, the surface free energy must be minimal with respect to several thermodynamic variables; the condition (delta F/ delta A) = 0 is not adequate by itself.

Measured work of deformation and repulsion of lecithin bilayers

We used three complementary techniques to vary the chemical potential of water in lipid/water mixtures; we measured the work of removing water from the multilayer lattice formed in water by the zwitterionic phospholipid egg lecithin. By x-ray diffraction, we observed the structural consequences of water removal. There are no discrete classes of "bound water" in this system; the work of removal is a continuous function of water content and lattice repeat spacing. From 30 to 3 A separation between bilayers there exists an exponential "hydration force" repulsion with a 2.6 A decay length. This interaction translates into a very large force to prevent contact between vesicles and planar membranes. It may be an important feature in controlling vesicle-to-cell fusion. As water is removed, bilayers not only move closer, but thicken as the lipid polar groups on the same bilayer move closer together. It is possible to divide the applied work into that of direct bilayer repulsion and that of bilayer deformation. We thus obtained a first determination of the lateral pressure required to create large increases in bilayer thickness and concomitant decreases in bilayer area. The lateral pressure reaches 25 dynes/cm for a 25% decrease in bilayer area. Systematic measurements of the mechanical properties of bilayers suffering such large deformation will allow critical tests of theories on bilayer stability and phase transition.

Energy transfer in lipid bilayers

The quenching of fluorescence due to energy transfer between a dilute, random array of donor and acceptor chromophores in lipid bilayer was measured and compared to theoretical expressions developed to predict the decrease in emission intensity under these circumstances. The observed intensity was found to be the same function of quencher concentration in both planar, multilamellar dispersions and small, spherical vesicles. The degree of quenching was accurately predicted by a simple relation derived in this paper, as well as a more complex equation previously developed by Tweet, et al. The results suggest that significant quenching may be observed even when the average donor-acceptor separation exceeds the Förster critical distance by severalfold. Application of these results to problems of current interest in membrane research are discussed.

Formation and properties of 1000-A-diameter, single-bilayer phospholipid vesicles

Two methods are reported for the formation of large, uniform-sized phospholipid vesicles. The methods involve the treatment of phospholipid, in the form of either small, sonicated vesicles or a dry lipid film, at a molar ratio of deoxycholate to phospholipid of 1:2. Subsequent removal of deoxycholate yields a stable preparation of vesicles. These vesicles are bounded by a single bilayer, have an average diameter of 1000 A, and are readily separated from sonicated vesicles (230 A) by gel filtration on Sepharose 4B. Since the 1000-A vesicles are capable of trapping enzymes and other macromolecules, they may prove valuable for the delivery of liposome-entrapped solutes to cells and for the localization of peptide segments of a spectrum of membrane-bound proteins.

Interaction of secreted nascent chains with surrounding membrane in Bacillus subtilis

To determine the length of secreted nascent polypeptide chain that is surrounded by membrane, we digested labeled nascent chains protruding from protoplasts of Bacillus subtilis with Pronase and isolated the residual ribosome-attached chains from the membrane-polysome fraction. Gel chromatography revealed a sharp major peak that had been protected by membrane plus bound ribosomes. The ribosomes themselves protected half as great a length. Because no free chain between the ribosome and the membrane was detected by Pronase treatment, the difference between the two protected lengths should measure the length protected by the membrane. More accurate measurements of these lengths, obtained by dansylation of the exposed NH2 terminus of the isolated fragments, yielded a difference of 21 amino acids. This value corresponds to an extended chain of 75 A, which is approximately the thickness of the bacterial cell membrane. We earlier presented evidence that bacterial ribosomes are attached to membrane solely by their secreted chain. The present results further show that after loss of the extracellular segment of the chain its attachment persists, at 37 degrees as well as 0 degrees C. These findings suggest that the chain does not slip through a passive membrane but is actively held within a channel.

Availability of lysophosphatidylcholine in single bilayer vesicles for hydrolysis by lysophospholipase

Single bilayer vesicles were prepared from total rat liver microsomal lipids to which 5 mol% lysophosphatidylcholine had been added. The availability of lysophosphatidylcholine for enzymatic hydrolysis by lysophospholipase (EC 3.1.1.5) was found to be higher in vesicles prepared by the cholate dispersion technique when compared with sonicated vesicles. Sepharose 4 B chromatography showed that the vesicles prepared by the cholate technique were smaller than those prepared by sonication. This is in contrast to previous observations for egg phosphatidylcholine vesicles. Total rat liver microsomal extracts were found to contain proteolipid, which could be removed by ether precipitation. Cholate vesicles prepared from proteolipid-free extracts were still smaller than sonicated vesicle from this extract. Experiments with [14C] dextran entrapped in the vesicles indicate that there is no loss of the permeability barrier of the vesicles for high molecular weight solutes during vesicle treatment with lysophospholipase. The high availability of lysophosphatidylcholine in cholate vesicles of total rat liver microsomal lipids is discussed in terms of a highly asymmetric distribution of lysophosphatidylcholine over the inner and outer monolayer of the bilayer.

Location and interactions of phospholipid and cholesterol in human low density lipoprotein from 31P nuclear magnetic resonance

The major phospholipids, phosphatidylcholine and spingomyelin, of low density lipoprotein (LDL) are accessible to small amounts of Pr3+, suggesting that the head groups of all mobile phospholipids are on the surface of the particle in contact with the aqueous medium. The major source of the nuclear Overhauser effect enhancement of 31P resonances is the N-methyl proton of the choline moiety, indicating close N-methyl phosphate group interactions, probably similar to those found previously in phospholipid vesicles. This behavior of the phospholipid head groups in LDL is similar to that in small vesicles without cholesterol, suggesting that in LDL most of the cholesterol is not associated with mobile, surface phospholipids. In contrast to LDL, where the presence of a large protein immobilizes some phospholipid head groups, immobilization does not occur in high density lipoprotein, consistent with occurrence of smaller peptides in the latter.

Morphology of egg phosphatidylcholine-cholesterole single-bilayer vesicles, studied by freeze-etch electron microscopy

Homogeneous, small, single-bilayer vesicles were prepared from egg phosphatidylcholine with various concentrations of cholesterol by ultrasonic dispersion in 0.1 M KCl, 0.01 M Tris, pH 8.0, buffer, followed by gel chromatography. The shape and size distributions of the fractionated vesicles were investigated for preparations with cholesterol compositions from 0 to 50 moles/100 moles, using freeze-etch electron microscopy. The size distribution was estimated from the shadow width of vesicles which were exposed by etching and the vesicle shape was checked by comparing the images obtained by tilting the replicas. The widths of the vesicle diameter distributions were relatively broad, corresponding to standard deviations in the range 60--90 A, but showing no systematic variation with cholesterol composition. In all cases it was found that 70% of the vesicle diameters lay within 150 A of the modal value. The apparent vesicle diameters remained constant for cholesterol compositions up to 20 moles/100 moles (modal diameter = 330 +/- 20 A, mean diameter = 350 +/- 3 A), but there was a sharp net increase in diameter at 30 moles cholesterol/100 moles reaching a model diameter of 430 +/- 20 A (mean diameter = 430 +/- 3 A) at 50 moles cholesterol/100 moles. Using the tilted microscope stage it was found that all vesicles were spherical at all cholesterol compositions studied, including those above 30 moles cholesterol/100 moles. The molecular mechanism by which cholesterol controls the vesicle size is discussed in terms of the asymmetric distribution of cholesterol across the vesicle bilayer.