Phospholipid topology and flip-flop in intestinal brush-border membrane.

The topological distribution of the two major phospholipids of brush-border membrane, phosphatidylcholine (PC) and phosphatidylethanolamine (PE), has been investigated using brush-border membrane vesicles from rabbit small intestine. Bee venom phospholipase A2 and phosphatidylcholine exchange protein from bovine liver were used as membrane probes. It is shown that the brush-border membrane retains its integrity under conditions of phospholipase hydrolysis and intermembrane phospholipid exchange. Kinetic analysis of the data of phospholipase hydrolysis and phospholipid exchange at temperatures under 10 degrees C shows that both PC and PE occur in two pools: a minor (about 25%) more readily accessible pool and a major one (about 75%) less readily available. The rate of PC exchange between these two pools is relatively fast. The half-time derived under conditions of phospholipase hydrolysis is of the order of 20 min. Under conditions of phospholipid exchange the exchange rates may be even faster. The difference in exchange kinetics observed with the two methods of probing is probably due to changes in membrane properties such as the bilayer fluidity induced by the probing process itself. It is proposed that the two pools represent the transverse distribution of the phospholipids. The two major phospholipids of brush-border membranes, PC and PE, would be distributed mainly on the inner (cytoplasmic) side of the brush-border membrane. The phospholipid exchange between the brush-border vesicles and unilamellar phosphatidylcholine vesicles in the presence of phosphatidylcholine exchange protein reveals that significant quantities of phospholipid are taken up by brush-border membrane independently, i.e., in a separate process independent of the exchange protein-catalyzed phosphatidylcholine exchange.

Lipid-cell interactions. A novel mechanism of transfer of liposome-entrapped substances into cells.

A new approach has been developed for studying the transfer of liposome-entrapped substances into cells. The cells are incubated with liposomes containing two markers that in the free (non-entrapped) state enter the cells at different rates. Comparison of the ratio of cell-associated markers applied either in free or in liposome-entrapped form permits the evaluation of different pathways of cellular uptake of the intraliposomal substances. When epithelial cell sheets were incubated with egg phosphatidylcholine liposomes containing two different sugars they became cell-associated at a ratio different from their initial ratio inside the liposomes. Since the cell-associated ratio was shifted towards the value observed when the cells were incubated with a mixture of the two sugars in the free state, it is suggested that the liposomes become permeable during incubation and that the liberated substances enter the cells in the free form. On the other hand, cell-liposome interaction was demonstrated by NMR measurement and gel-filtration experiments to result in transformation of small unilamellar liposomes into larger multilayered aggregates. This transformation depends on the contact of the liposomes with the cell sheet. It is supposed that interliposomal aggregation is the underlying mechanism of cell-induced leakage of liposomes.

Affinity chromatography of the phosphatidylcholine exchange protein from bovine liver.

Affinity chromatography has been used to purify the phosphatidylcholine exchange protein from bovine liver. The affinity resin consisted of 1-acyl-2-(9-carboxy)nonyl-glycero-3-phosphocholine linked to AH-Sepharose 4 B via the carboxyl group. Application of a crude exchange protein fraction to the affinity column resulted in a complete adsorption of the phosphatidylcholine exchange protein. The exchange protein eluted with a buffer containing 0.15% sodium deoxycholate. The most active fraction was 130-fold purified and accounted for 62% of the activity.

Diol lipids. Acylated ethyleneglycol glycosides. A new type of natural glycolipid.

A new type of glycolipid has been detected in ripening corn seeds. The presence of ethyleneglycol, galactose, glucose and fatty acids was shown by degradation studies. The products of alkaline deacylation were identified as galactosyl- and glycosylethyleneglycol by thin-layer chromatography and combined gas-liquid chromatography-mass spectrometry. The native ethyleneglycol galactolipid was isolated by distribution of the total lipids between heptane and 95% methanol, following silica gel column chromatography of the methanol soluble fraction. Analysis of the alkaline deacylation products of the isolated ethyleneglycol lipid as well as examination of the mass-spectrum of its tetraacetate showed the new lipid to have the structure of 1-acyl-2-(O-beta-D-galacto-pyranosyl)ethyleneglycol with palmitic, oleic and stearic acids as the main fatty acid components. The fragmentation patterns under electron impact of trimethylsilyl ethers of synthetic ethyleneglycol glycopyranosides and of 1-palmitoyl-2-(O-beta-D-tetraacetylgalactopyranosyl)ethyleneglycol are described.

Investigation of the inside-outside distribution, intermembrane exchange and transbilayer movement of phospholipids in sonicated vesicles by shift reagent NMR.

1. A new NMR approach is described for the investigation of transbilayer asymmetry in phospholipid vesicles consisting of phosphatidylcholine and negatively charged phospholipids. The method makes use of the dependence of the psuedocontact shift of the N-methyl proton resonance induced by paramagnetic ions on the surface concentration of negatively charged phospholipids. When two differently shifting paramagnetic probes are applied from the outside and the inside of a vesicular membrane the transbilayer phospholipid distribution can be estimated without knowledge of the inner and outer radii of the vesicles and the packing density of the phospholipid molecules. 2. The method was employed to study the transbilayer asymmetry in vesicles obtained by cosonication of phosphatidylcholine with phosphatidylserine, phosphatidylglycerol or phosphatidylinositol. The three negative phospholipids were found to distribute with a higher surface concentration in the inner vesicular shell than in the outer one when their total content did not exceed 25 mol%. However, as the amount of negatively charged phospholipids increases the ratio of their inside to outside surface concentrations, i.e., the transbilayer asymmetry of the vesicles, decreases. Prolonged incubation (for several days) does not change the compositional asymmetry of the cosonicated vesicles. 3. By the 'double-probe' technique it was established that spontaneous exchange between separately sonicated phosphatidylcholine and phosphatidylinositol vesicles results in formation of highly asymmetric mixed vesicles with phosphatidylinositol residing only in the outer monolayer. In the presence of antioxidant (alpha-tocopherol) the bilayer asymmetry is preserved for days. However lipid peroxidation induces rapid transbilayer movement (flip-flop) of phospholipids leading to an 'inverted' asymmetry resembling that of cosonicated vesicles. It is suggested that lipid peroxidation promotes phospholipid flip-flop by partially converting the bilayer structure into a non-bilayer configuration. Moderate quantities of lysophosphatidylcholine (up to 15 mol%) induce neither detectable perturbations of the bilayer nor rapid phospholipid flip-flop.

The interaction of prostaglandins with high-density lipoproteins: a non-equilibrium model of ligand-receptor interaction.

Using high-density lipoproteins (HDL) labeled with a fluorescent phospholipid probe (an anthrylvinyl-labeled analogue of sphingomyelin) it was found that low amounts (10(-12) M) of the prostaglandins E1 and F2 alpha induced different structural changes of the HDL surface, whereas prostaglandin E2 had no effect. The effects of prostaglandin E1 on HDL were largely paralleled by those of this prostaglandin on synthetic recombinants prepared from apolipoprotein A1, phospholipids and cholesterol. The prostaglandin E1-HDL interaction resembled that of a ligand with a receptor site because it was specific, reversible, concentration- and temperature-dependent and saturable. However, the maximal HDL retaining capacity for prostaglandin E1 as determined by equilibrium dialysis was very low, and a single prostaglandin E1 molecule was able to induce structural changes in a large number of discrete lipoprotein particles. To explain this remarkable fact, a non-equilibrium model of ligand-receptor interaction is proposed. According to this model in open systems characterized by a short life-time of the ligand-receptor complex, high diffusion rates of the ligand and long relaxation times which exceed the interval between two successive ligand-receptor occupations, the ligand-induced changes will accumulate, resulting in amplification of the primary biological signal. It is emphasized that the low mobility of lipids constituting the environment of the receptor protein plays a critical role in this type of signal amplification.

The molecular organization of the influenza virus surface. Studies using photoreactive and fluorescent labeled phospholipid probes.

The membrane structures of remantadin-sensitive and remantadin-resistant influenza virus strains were studied using a photoreactive fatty acid as well as analogues of phosphatidylcholine, phosphatidylethanolamine and sphingomyelin, carrying a fluorescent or photoreactive reporter group at the end of one of the aliphatic chains. The results obtained demonstrated for the first time that the phospholipids of the viral membrane form lateral domains differing by the fluidity of their hydrocarbon chains and, probably, by the head-group composition of the lipids. The hemagglutinin small subunit (HA2) was shown to protrude into the apolar region of the phospholipid bilayer, whereas the M1 protein makes contact only with the inner surface. In the remantadin-sensitive virions the heavy hemagglutinin chain (HA1) appears not to be in contact with the lipid bilayer, whereas in the remantadin-resistant strain HA1 has a hydrophobic segment that proved to be inserted into the bilayer.

Lipid-cell interactions. Liposome adsorption and cell-to-liposome lipid transfer are mediated by the same cell-surface sites.

The competitive behavior of solid vs. fluid liposomes in liposome-to-cell adsorption and cell-to-liposome lipid transfer processes was investigated with L cells and FBT epithelial sheets. Binding, transfer and 31P-NMR experiments have demonstrated that: (i) solid liposomes adhere to the cell surface as integral vesicles retaining the entrapped substances; (ii) fluid liposomes are partly disintegrated at the cell surface with concomitant entry of entrapped substances into the cytoplasm, while their lipids remain on the cell surface; (iii) fluid liposomes that escape lysis dissociate from the cell, taking away cell lipid molecules. The latter process underlies the mechanism of cell-to-fluid liposome lipid transfer. In contrast, no lipid transfer occurs between the plasma membrane and solid liposomes. Cell-bound solid liposomes interfere with the transfer of cell lipids to fluid liposomes, while these in turn inhibit the binding of solid liposomes to the cell surface. Moreover, cell-induced aggregation of both fluid and solid freshly added liposomes is also inhibited by preincubation of the cells with either solid or fluid liposomes. Thus, different types of interaction of both fluid and solid liposomes with the cell are mediated by the same (or closely related) sites on the cell surface.

Stimulation of platelet adhesion and activation by ganglioside GD3 adsorbed to plastic.

Platelet interaction with gangliosides GD3, GM3, GM1, GD1a and GT1b has been investigated. These gangliosides were previously identified in the vessel wall and ganglioside GD3 was found to accumulate selectively in the intima of atherosclerotic vessels. Gangliosides were adsorbed to plastic and incubated with 51Cr-labeled platelets. The adhesion of gel-filtered platelets to ganglioside GD3 was 3-4 times higher than to other immobilized gangliosides and to albumin-treated plastic. As was shown by scanning electron microscopy, GD3 stimulated intensive spreading of adherent platelets and formation of surface-bound aggregates, while only single unspread platelets were present on the surfaces coated with other gangliosides. GD3 isolated from milk and from human aorta possess the same stimulating activity. Platelet adhesion to GD3 decreased significantly in the presence of the stable prostacyclin analogue, carbacyclin.

Influence of cholesterol and prostaglandin E1 on the molecular organization of phospholipids in the erythrocyte membrane. A fluorescent polarization study with lipid-specific probes.

Anthryl-labeled fluorescent probes closely mimicking phosphatidylcholine and sphingomyelin were applied to study the state of these phospholipids in the rabbit erythrocyte membrane. At normal cholesterol levels both probes exhibited higher fluorescence polarization values in the membranes than in phospholipid vesicles of similar lipid composition, indicating a decreased fluidity of the probe environment in erythrocyte ghosts. In ghosts prepared from normal erythrocytes no evidence of lateral separation of phosphatidylcholine and sphingomyelin was found. At higher cholesterol levels, however, these lipids appear to segregate. Probably the effect of cholesterol on the erythrocyte membrane lipids involves lipid-protein interactions. At physiological concentrations, prostaglandin E1 only weakly affects the state of phosphatidylcholine and sphingomyelin in erythrocyte membranes. Cholesterol enrichment amplifies the effect of prostaglandin E1. Although the prostaglandin E1-induced changes depended much upon whether the ghosts were enriched with cholesterol in vitro or in vivo, with both types of ghosts effects of prostaglandin E1 were seen at extremely low effector concentrations that may have presented a few molecules of prostaglandin per ghost. The structural and functional significance of these findings is discussed.

Interaction of low-density lipoproteins with gangliosides.

The ganglioside uptake capacity of human serum low-density lipoproteins (LDL), the mode of ganglioside-LDL binding, and the influence of gangliosides on the floatation properties, size distribution, stability and fluorescence of LDL were investigated. The data obtained suggest that both hydrophobic and electrostatic forces are involved in formation of ganglioside-LDL complexes, but the former appear to be more important. Although association of gangliosides with LDL is predominantly unspecific, nonsaturable, and weak, a small saturable component due to specific ganglioside-apolipoprotein binding, also appears to be involved. In the presence of gangliosides the lipoprotein particles aggregate, the intrinsic fluorescence of LDL and their interaction with antibodies against apo-B change indicating that the state of apo-B [corrected] is modified by gangliosides.

The interaction of prostaglandins with serum low-density lipoproteins.

The interaction of human serum low-density lipoproteins (LDL) with various types of prostaglandins (PG) was studied using equilibrium dialysis, steady-state fluorescence polarization spectroscopy and photolabeling methods. Low concentrations (10(-13)-10(-9) M) of PGE1 and PGF2 alpha were shown to induce specific rearrangements of the lipids on the LDL surface, whereas the closely related PGE2 and PGF1 alpha had no effect. With fluorescent labeled LDL, the PGE1-induced changes of the steady-state fluorescence polarization (P) were shown to be time- and concentration-dependent, saturable and reversible. However, equilibrium dialysis revealed a very low binding capacity of LDL for PGE1 (approx. 1 prostaglandin molecule per 600 LDL particles). Approximately the same PGE1 concentration was sufficient to cause maximal changes of P, to enhance the binding to apolipoprotein B of a photoreactive sphingomyelin analogue inserted into the LDL surface and to alter the thermal phase behavior of the LDL surface lipids. It is proposed that the LDL surface rearrangement caused by prostaglandins is due to the interaction of prostaglandins with apolipoprotein B, resulting in formation of short-lived complexes. The mechanism of this interaction is discussed in terms of the non-equilibrium ligand-receptor interaction model proposed earlier to explain the interaction of prostaglandins with high-density lipoproteins (Bergelson, L.D. et al. (1987) Biochim. Biophys. Acta 921, 182-190). It is suggested that direct prostaglandin-lipoprotein interactions may play a role in the homeostasis of cholesterol.

Influence of platelet activation factor and prostaglandins on cholesterol esterification in human plasma.

Prostaglandin E1 (PGE1) and platelet activation factor (PAF) were found respectively to activate or to inhibit cholesterol esterification in whole plasma but not in lipoprotein-deficient plasma. It is suggested that these effects are mediated by the interaction of PGE1 and PAF with high-density lipoproteins [(1984) FEBS Lett. 173, 291-294]. Possible physiological implications of these findings are discussed.

The binding of the B-chain of ricin to Burkitt lymphoma cells. A new approach to ligand-receptor interaction studies.

It is shown that conformational changes of receptor proteins brought about by binding of a ligand induce changes in the lipid environment of the receptor that can be monitored by fluorescent lipid probes. On this basis a new approach to studies of ligand-receptor binding is proposed. Using the interaction of the ricin B-chain with Burkitt lymphoma cells as an example and fluorescent labelled sphingomyelin as a probe, the ligand-induced changes of fluorescence anisotropy were shown to be concentration-dependent and to permit determination of the binding constant and the number of receptor-binding sites. The method was found to be specific and highly sensitive, allowing detection of the action of one RB molecule per cell. Scatchard analysis of the binding of 125I-RB demonstrated the presence on the cell surface of two binding sites with Kd approximately 10(-10) and approximately 10(-8) M, respectively. Only the high-affinity sites were detected by the fluorescence technique. Saturation of these sites resulted in maximum inhibition of protein synthesis.

Diol lipids are activators of protein kinase C.

Diol lipids (dioleoyl- and dioctanoylethylene glycol) at relatively low concentrations (approximately 10 microM) were found to activate significantly protein kinase C in the presence of phosphatidylserine or phosphatidylinositol. Since diol lipids are widespread minor lipid constituents of many cells [(1974) Chem. Ind., 597-604], it has been suggested that they may be involved in the maintaining of basal protein kinase C activity in the absence of external stimuli.

Low-density lipoproteins interact with liposome-binding sites on the cell surface.

Under physiological conditions significant amounts of low-density lipoprotein LDL particles ar taken up by cells independently of specific high-affinity LDL receptors (apo-B receptors). Previously it was established that some cells contain surface sites capable of binding liposomes. We proposed that liposome-binding sites could contribute to LDL interaction with the cell surface via phospholipid molecules of LDL particles. To check this hypothesis we studied the competitive interaction of human LDL and DPPC liposomes with mouse embryo fibroblasts depleted of apo-B receptors by preliminary incubation with LDL. We have found that after removal of the liposome-binding sites from cell lamellae these areas of the cell surface lose their ability to bind LDL.

Binding of specific ligands to muscarinic receptors alters the fluidity of membrane fragments from rat brain. A fluorescence polarization study with lipid-specific probes.

The previously suggested method of following ligand-receptor interactions by measuring ligand-induced changes in membrane fluidity [(1986) FEBS Lett. 194, 313-316] was employed to study the binding of specific ligands of the muscarinic receptor to rat brain membrane fragments containing a fluorescent analogue of phosphatidylcholine (APC) as a membrane probe. Upon addition of carbachol and atropine in low concentrations the fluorescence polarization of the APC-labeled membranes decreased significantly demonstrating that binding of these ligands to the muscarinic receptor increases the fluidity of its lipid environment. The fluidity changes were specific, concentration-dependent and saturable. In comparison with radioligand assays the fluorescent lipid probe method proved to be much more sensitive but the Kd values obtained by the two methods differed considerably.

Influence of prostaglandins on the lipid transfer between human high density and low density lipoproteins.

Prostaglandin (PG) E1 was demonstrated to stimulate the transfer of phosphatidylcholine and cholesterol esters from human high density lipoproteins (HDL3) to low density lipoproteins (LDL). The enhancement effect of PGE1 on the interlipoprotein lipid transfer was seen at low PG concentrations under conditions of spontaneous exchange as well as in the presence of lipoprotein-depleted plasma, or partly purified plasma lipid exchange protein. PGE2 and PGF2 alpha showed no significant influence on the interlipoprotein lipid transfer. Evidence is presented suggesting that the PGE1-induced stimulation of interlipoprotein lipid exchange results in enhancement of LCAT-catalyzed cholesterol esterification in plasma. It is proposed that the effect of PGE1 is due to the previously described PGE1-induced reorganization of the HDL surface [(1984) FEBS Lett. 173, 291-293] and that PG-lipoprotein interaction may be a factor regulating cholesterol homeostasis.

Action of influenza virus neuraminidase on gangliosides. Haemagglutinin inhibits viral neuraminidase.

The action of partly purified neuraminidase (NA) of influenza A virus, a mixture of detergent solubilized NA and haemagglutinin (HA) and of intact virions on gangliosides GT1b, GD1a, GD1b, GM1 was studied. The viral NA transformed GT1b mainly into GD1b with formation of only minor amounts of GM1. HA was found to inhibit the hydrolysis activity of viral NA. At the same time viral NA transformed GD1a quantitatively into GM1 which was not hydrolyzed by the enzyme. These results suggest that the function of NA is to transfer the 'primary' receptor (such as GT1b) into the proper carbohydrate sequence (GD1b-like) which is proposed to serve as the minimal structure required for influenza virus reception.

Interaction of prostaglandin E1 with human high density lipoproteins.

Prostaglandin E1 has been shown to interact with serum high density lipoproteins (HDL) in a manner resembling the interaction of a ligand with a high affinity binding site. The presence of 10(-12)-10(-10) M prostaglandin E1 induces a rearrangement of the HDL surface lipids and probably influences the biological functions of the lipoproteins.