Direct inhibition of phospholipid scrambling activity in erythrocytes by potassium ions.

The exposure of phosphatidylserine (PS) at the cell surface plays a critical role in blood coagulation and serves as a macrophage recognition moiety for the engulfment of apoptotic cells. Previous observations have shown that a high extracellular [K(+)] and selective K(+) channel blockers inhibit PS exposure in platelets and erythrocytes. Here we show that the rate of PS exposure in erythrocytes decreases by approximately 50% when the intracellular [K(+)] increases from 0 to physiological concentrations. Using resealed erythrocyte membranes, we further show that lipid scrambling is inducible by raising the intracellular [Ca(2+)] and that K(+) ions have a direct inhibitory effect on this process. Lipid scrambling in resealed ghosts occurs in the absence of cell shrinkage and microvesicle formation, processes that are generally attributed to Ca(2+)-induced lipid scrambling in intact erythrocytes. Thus, opening of Ca(2+)-sensitive K(+) channels causes loss of intracellular K(+) that results in reduced intrinsic inhibitory effect of these ions on scramblase activity.

Uncoupling of the membrane skeleton from the lipid bilayer. The cause of accelerated phospholipid flip-flop leading to an enhanced procoagulant activity of sickled cells.

We have previously reported that the normal membrane phospholipid organization is altered in sickled erythrocytes. More recently, we presented evidence of enhanced transbilayer movement of phosphatidylcholine (PC) in deoxygenated reversibly sickled cells (RSC) and put forward the hypothesis that these abnormalities in phospholipid organization are confined to the characteristic protrusions of these cells. To test this hypothesis, we studied the free spicules released from RSC by repeated sickling and unsickling as well as the remnant despiculated cells. The rate of transbilayer movement of PC in the membrane of deoxygenated remnant despiculated cells was determined by following the fate of 14C-labelled PC, previously introduced into the outer monolayer under fully oxygenated conditions using a PC-specific phospholipid exchange protein from beef liver. The rate of transbilayer movement of PC in the remnant despiculated cells was significantly slower than in deoxygenated native RSC and was not very much different from that in oxygenated native RSC or irreversibly sickled cells. The free spicules had the same lipid composition as the native cells, but were deficient in spectrin. These spicules markedly enhanced the rate of thrombin formation in the presence of purified prothrombinase (Factor Xa, Factor Va, and Ca2+) and prothrombin, indicating the exposure of a significant fraction of phosphatidylserine (PS) in the outer monolayer. This effect was not observed when the spicules in this assay were replaced by normal erythrocytes, deoxygenated native RSC, or a deoxygenated sample of RSC after repetitive sickling/unsickling. The results are interpreted to indicate that the destabilization of the lipid bilayer in sickled cells, expressed by the enhanced flip-flop of PC and the exposure of PS in the outer monolayer, occurs predominantly in those parts of the membrane that are in spicular form.

The enzymatic synthesis of phosphatidylserine and purification by CM-cellulose column chromatography.

Phosphatidylserine has been prepared from phosphatidylcholine by a one-step transphosphatidylation catalyzed by phospholipase D in the presence of L-serine. The resulting mixture of phosphatidylserine and phosphatidic acid is easily and rapidly separated by CM-cellulose column chromatography using step=wise elution with solvents containing increasing percentages of methanol in chloroform. The over-all yield of the procedure is 40-50% depending on the scale of the preparation. CM-Cellulose column chromatography proved to be extremely useful in separating phospholipid mixtures obtained by phosphatidyltransferase reactions of phospholipase D and is also suitable for fractionation of other lipid extracts.

Interaction between phosphatidylserine and the isolated cytoskeleton of human blood platelets.

Binding experiments were performed to demonstrate a direct interaction between cytoskeletons from human blood platelets and phosphatidylserine. A centrifugation technique using radiolabeled phosphatidylserine-vesicles and Triton X-100 insoluble residues from unstimulated human platelets was used to assess the binding. Interaction between cytoskeleton and phospholipid is demonstrated to be specific for phosphatidylserine. No binding was observed for phosphatidylcholine. The binding of phosphatidylserine was saturable and dependent on the concentration of cytoskeleton used. The interaction between phosphatidylserine and the cytoskeleton appeared to be completely reversible. The existence of a reversible and specific interaction between phosphatidylserine and the cytoskeleton of unstimulated platelets would suggest a role for the cytoskeleton in the maintenance of the asymmetric distribution of this lipid in the plasma membrane. We have previously shown (Comfurius et al. (1985) Biochim. Biophys. Acta 815, 143-148) that in activated platelets a strong correlation exists between degradation of platelet cytoskeletal proteins by the endogenous calcium-dependent proteinase (calpain) and exposure of phosphatidylserine at their outer surface. Nevertheless, hydrolysis of the isolated cytoskeleton by calpain did not result in a change in the parameters of the binding between phosphatidylserine and cytoskeleton. Also, sulfhydryl oxidation of the cytoskeleton by diamide did not affect its binding properties for phosphatidylserine, in spite of the fact that diamide treatment of platelets results in exposure of phosphatidylserine at the outer surface. Exposition of phosphatidylserine upon activation of platelets cannot be directly ascribed to a change in affinity or number of binding sites of the modified cytoskeleton as measured in model systems. However, it cannot be excluded that topological rearrangements of the cytoskeleton as occur within the cell during platelet activation lead to a decreased contact between cytoskeleton and lipid, irrespective of the binding parameters.

The involvement of cytoskeleton in the regulation of transbilayer movement of phospholipids in human blood platelets.

Activation of human platelets by different activators resulted in a different extent of degradation of the cytoskeletal proteins actin-binding protein and myosin, as well as of the non-cytoskeletal protein P235. The highest extent of proteolysis was observed with Ca-ionophore A23187 and decreased on going from A23187 greater than collagen plus thrombin greater than collagen greater than thrombin = ADP. The same order of potency has been found previously ((1983) Biochim. Biophys. Acta 736, 57-66) for the ability of platelet activators to induce exposure of aminophospholipids in the outer leaflet of the platelet plasma membrane, and to stimulate platelets to become procoagulant. Degradation of cytoskeletal proteins as a result of platelet stimulation by collagen plus thrombin was prevented in the presence of dibutyryl cAMP or EDTA but not in the presence of aspirin. This also runs in parallel with platelet procoagulant activity. Moreover, platelets from a patient with a partial deficiency in platelet procoagulant activity revealed a diminished extent of degradation of cytoskeletal proteins upon platelet stimulation with collagen plus thrombin. It is concluded that alterations in cytoskeletal organization upon platelet stimulation may lead to alterations in the orientation of (amino)phospholipids in the plasma membrane, and may therefore play a regulatory role in the expression of platelet procoagulant activity.

Lipid-protein interactions in blood coagulation.

It has long been appreciated that lipids, particularly anionic phospholipids, promote blood coagulation. The last two decades have seen an increasing insight into the kinetic and mechanistic aspects regarding the mode of action of phospholipids in blood coagulation. This essay attempts to review these developments with particular emphasis on the structure of lipid-binding domains of blood coagulation proteins, and the variable effect of phospholipid composition on the interaction with these proteins. Some examples are discussed of how lipid membranes direct the pathway of enzymatic conversions in blood coagulation complexes, also illustrating that the membrane lipid surface is more than an inert platform for the assembly of coagulation factors. Finally, the controlled exposure of procoagulant lipid on the surface of blood cells is shortly reviewed, and an example is discussed of how interference with lipid-protein interactions in blood coagulation may result in pathological phenomena.

Changes in membrane phospholipid distribution during platelet activation.

Exposure of phospholipids at the outer surface of activated and control platelets was studied by incubation with a mixture of phospholipase A2 from Naja naja and bee venom, solely or in combination with sphingomyelinase from Staphylococcus aureus, using conditions under which cell lysis remained below 10%. Incubation with phospholipase A2 alone revealed a markedly increased susceptibility of the phospholipids in platelets activated by a mixture of collagen plus thrombin, by the SH-oxidizing compound diamide, or by calcium ionophore A23187, as compared to control platelets or platelets activated separately by collagen or thrombin. Collagen plus thrombin, diamide, and ionophore treated platelets revealed an increased exposure of phosphatidylserine at the outer surface accompanied by a decreased exposure of sphingomyelin, as could be concluded from incubations with a combination of phospholipase A2 and sphingomyelinase. These alterations were much less apparent in platelets activated either by thrombin or by collagen alone. The increased exposure of phosphatidylserine in activated platelets is accompanied by an increased ability of the platelets to enhance the conversion of prothrombin to thrombin by coagulation factor Xa, in the presence of factor Va and calcium. It is concluded that the altered orientation of the phospholipids in the plasma membrane of platelets activated by collagen plus thrombin, by diamide, or by calcium ionophore, is the result of a transbilayer movement. Moreover, the increased exposure of phosphatidylserine in platelets stimulated by the combined action of collagen and thrombin might be of considerable importance for the hemostatic process.

Fluoride-dependent calcium-induced platelet procoagulant activity shows that calpain is involved in increased phospholipid transbilayer movement.

Treatment of platelets with fluoride (10 mM) was found to result in a transient increase in Ca2+-permeability of the platelet plasma membrane. This phenomenon was used to provide supplementary evidence for the suggestions made earlier (Comfurius et al. (1985) Biochim. Biophys. Acta 815, 143; Verhallen et al. (1987) Biochim. Biophys. Acta 903, 206), that cytoskeletal disrupture by calpain is involved in the process leading to transbilayer movement of phosphatidylserine during expression of platelet procoagulant activity. This was achieved by relating both calpain activity and exposure of phosphatidylserine with platelet procoagulant activity. It was found that only upon addition of extracellular Ca2+ to fluoride-treated platelets, procoagulant activity, expressed as prothrombinase activity, and calpain activity, estimated from protein patterns after gel electrophoresis, were generated. Both Ca2+-inducible prothrombinase activity and calpain activity followed an identical time-course during incubation with fluoride: after a time-lag of about 10 min they sharply increased towards a peak level. Upon further incubation with fluoride, both activities decreased towards a final plateau, still above basal level. The presence of leupeptin during incubation with fluoride was found to inhibit Ca2+-inducible calpain activity and prothrombinase activity in an identical way. Ca2+-inducible exposure of phosphatidylserine, as determined with extracellular phospholipase A2, showed a similar pattern as Ca2+-inducible calpain activity and prothrombinase activity. From the strict parallelism between prothrombinase activity, calpain activity and exposure of phosphatidylserine, it is concluded that calpain plays an important role in the activation-dependent transbilayer movement of phosphatidylserine during expression of platelet procoagulant activity. It is suggested that degradation of the platelet membrane-skeleton by calpain disturbs the structural organization of the lipid bilayer of the platelet plasma membrane leading to enhanced transbilayer movement of phospholipids and appearance of phosphatidylserine at the platelet outer surface.

Correlation between calpain-mediated cytoskeletal degradation and expression of platelet procoagulant activity. A role for the platelet membrane-skeleton in the regulation of membrane lipid asymmetry?

The relationship between platelet calpain-activity and platelet procoagulant-activity was investigated by comparison of the time course of their generation after platelet stimulation by calcium ionophore A23187, or by the combined action of collagen and thrombin, or during exposure of platelets to the local anesthetics dibucaine or tetracaine. In addition, the Ca2+ dose-response curves of both activities in intact platelets, obtained by stimulation with A23187 in the presence of Ca2+/HEDTA-buffers, were compared. Platelet procoagulant activity was determined by assaying for prothrombinase activity in the presence of saturating concentrations of factors Xa, Va, and prothrombin. Platelet calpain activity was monitored by the degradation of its major substrates (filamin, talin, myosin) and the formation of their fragments as judged from protein patterns after gel electrophoresis. Platelet stimulation by A23187 resulted in a fast increase in prothrombinase activity, reaching its maximum level after about 20 seconds. Filamin and talin were completely hydrolysed within 15 s, and myosin was partly degraded between 15 and 30 s after platelet activation. When platelets were activated by collagen plus thrombin, prothrombinase activity was generated with a sigmoid time course, the steepest increase being observed between 1 and 2 min after platelet activation. Proteolysis of filamin and talin occurred between 0.5 and 1.5 min after platelet activation, while degradation of myosin became visible after 2 to 2.5 min. Dibucaine and tetracaine were both found to be potent stimulators of prothrombinase activity, with half-maximal activities obtained at 0.7 and 2.8 mM, respectively. Using suboptimal concentrations of both local anesthetics, it was found that the generation of prothrombinase activity closely paralleled that of calpain activity over a time course of 1 hour. Ca2+ titration of intact platelets using A23187 and Ca2+/HEDTA buffers, revealed half-maximal response at about 15 microM free Ca2+ for both calpain and prothrombinase activity. These findings strongly suggest a causal relationship between generation of a procoagulant platelet surface and calpain-mediated degradation of filamin, talin, and myosin. Since an increased procoagulant activity reflects an increased exposure of phosphatidylserine at the platelet outer surface, the present findings suggest that platelet cytoskeletal proteins are involved in the regulation of membrane lipid asymmetry.

Calcium-induced transbilayer scrambling of fluorescent phospholipid analogs in platelets and erythrocytes.

The non-random distribution of phospholipids in the plasma membrane of human platelets and erythrocytes is at least partially maintained by the ATP-dependent aminophospholipid translocase, but can be disturbed by a calcium-induced scrambling of lipids. Using fluorescent NBD-phospholipid analogs, we demonstrate that in both cells the aminophospholipid translocase has a slightly higher preference for the naturally occurring L-isomer of the polar headgroup of phosphatidylserine as compared to the D-isomer. Calcium-induced outward movement of internalized phosphatidylserine probe, however, is not affected by the stereochemical configuration of the serine headgroup and is virtually identical to both the inward and outward movement of the phosphatidylcholine probe. The data also indicate that both in platelets and red blood cells the calcium-induced transbilayer movement is bidirectional and involves all major phospholipid classes, with reorientation rates of sphingomyelin being appreciably lower than that of the other phospholipid classes. While our results largely support earlier observations on red cells, they clearly differ from a recent study on platelets which suggested that calcium-induced scrambling is restricted to aminophospholipids and would not involve cholinephospholipids. The present results indicate that the same mechanism is responsible for calcium-induced lipid scrambling in red blood cells and platelets.

Lipid translocation across the plasma membrane of mammalian cells.

The plasma membrane, which forms the physical barrier between the intra- and extracellular milieu, plays a pivotal role in the communication of cells with their environment. Exchanging metabolites, transferring signals and providing a platform for the assembly of multi-protein complexes are a few of the major functions of the plasma membrane, each of which requires participation of specific membrane proteins and/or lipids. It is therefore not surprising that the two leaflets of the membrane bilayer each have their specific lipid composition. Although membrane lipid asymmetry has been known for many years, the mechanisms for maintaining or regulating the transbilayer lipid distribution are still not completely understood. Three major players have been presented over the past years: (1) an inward-directed pump specific for phosphatidylserine and phosphatidylethanolamine, known as aminophospholipid translocase; (2) an outward-directed pump referred to as 'floppase' with little selectivity for the polar headgroup of the phospholipid, but whose actual participation in transport of endogenous lipids has not been well established; and (3) a lipid scramblase, which facilitates bi-directional migration across the bilayer of all phospholipid classes, independent of the polar headgroup. Whereas a concerted action of aminophospholipid translocase and floppase could, in principle, account for the maintenance of lipid asymmetry in quiescent cells, activation of the scramblase and concomitant inhibition of the aminophospholipid translocase causes a collapse of lipid asymmetry, manifested by exposure of phosphatidylserine on the cell surface. In this article, each of these transporters will be discussed, and their physiological importance will be illustrated by the Scott syndrome, a bleeding disorder caused by impaired lipid scrambling. Finally, phosphatidylserine exposure during apoptosis will be briefly discussed in relation to inhibition of translocase and simultaneous activation of scramblase.

Organization of phospholipids in human red cell membranes as detected by the action of various purified phospholipases.

1. The action of eight purified phospholipases on intact human erythrocytes has been investigated. Four enzymes, e.g. phospholipases A2 from pancreas and Crotalus adamanteus, phospholipase C from Bacillus cereus, and phospholipase D from cabbage produce neither haemolysis nor hydrolysis of phospholipids in intact cells. On the other hand, both phospholipases A2 from bee venom and Naja naja cause a non-haemolytic breakdown of more than 50% of the lecithin, while sphingomyelinase C from Staphylococcus aureus is able to produce a non-lytic degradation of more than 80% of the sphingomyelin. 2. Phospholipase C from Clostridium welchii appeared to be the only lipolytic enzyme tested, which produces haemolysis of human erythrocytes. Evidence is presented that the unique properties of the enzyme itself, rather than possible contaminations in the purified preparation, are responsible for the observed haemolytic effect. 3. With non-sealed ghosts, all phospholipases produce essentially complete breakdown of those phospholipids which can be considered as proper substrates for the enzymes involved. 4. Due to its absolute requirement for Ca2+, pancreatic phospholipase A2 can be trapped inside resealed ghosts in the presence of EDTA, without producing phospholipid breakdown during the resealing procedure. Subsequent addition of Ca2+ stimulates phospholipase A2 activity at the inside of the resealed cell, eventually leading to lysis. Before lysis occurs, however, 25% of the lecithin, half of the phosphatidylethanolamine and some 65% of the phosphatidylserine can be hydrolysed. This observation is explained in relation to an asymmetric phospholipid distribution in red cell membranes.

Structural and functional characterization of a prothrombin activator from the venom of Bothrops neuwiedi.

A prothrombin activator from the venom of Bothrops neuwiedi was purified by gel filtration on Sephadex G-100, ion-exchange chromatography on DEAE-Sephacel and affinity chromatography on a Zn2+-chelate column. The overall purification was about 200-fold, which indicates that the prothrombin activator comprises about 0.5% of the crude venom. The venom activator is a single-chain protein with an apparent molecular weight of 60 kDa. It readily activated bovine prothrombin with a Km of 38 microM and a Vmax of 120 mumol prothrombin activated per min per mg of venom activator. Venom-catalyzed prothrombin activation was not accelerated by the so-called accessory components of the prothrombinase complex, phospholipids plus Ca2+ and Factor Va. Gel-electrophoretic analysis of prothrombin activation indicated that the venom activator only cleaved the Arg-323-Ile-324 bond of bovine prothrombin, since meizothrombin was the only product of prothrombin activation. The activator did not hydrolyze commercially available p-nitroanilide substrates and its prothrombin-converting activity was not inhibited by benzamidine, phenylmethylsulfonyl fluoride, dansyl-Glu-Gly-Arg-chloromethyl ketone and soy-bean trypsin inhibitor. However, chelating agents such as EDTA, EGTA and o-phenanthroline rapidly destroyed the enzymatic activity of the venom activator. The activity of chelator-treated venom activator could be partially restored by the addition of an excess CaCl2. These results indicate that the venom activator remarkably differs from Factor Xa and that the enzyme is not a serine proteinase, but likely belongs to the metalloproteinases. The structural and functional properties of the venom prothrombin activator from B. neuwiedi are similar to those reported for the venom activator from Echis carinatus.

Loss of phospholipid asymmetry in dilauroylphosphatidylcholine induced plasma membrane vesicles from human platelets.

Incubation of human platelets with unilamellar vesicles composed of dilauroylphosphatidylcholine (DLPC) induces shedding of small vesicular structures from the platelet plasma membrane. No significant cell lysis is observed during the process of shedding. Isolated spicules contain the major membrane glycoproteins, Ib, IIb, and IIIa, which are used to define the sidedness of the spicule membrane. These glycoproteins are completely susceptible to chymotrypsin treatment, whereas cytoskeletal proteins are inaccessible towards this enzyme. This demonstrates that the spicule membranes have a right-side-out orientation in as far as membrane proteins are concerned. Isolated spicules were 30-fold more active than platelets in stimulating prothrombin conversion to thrombin by the prothrombinase complex (factors Xa, Va and Ca2+). The increased prothrombinase activity reflects an increased amount of phosphatidylserine in the outer leaflet of the spicule membrane. Protein analysis of platelet spicules and native platelets reveals a number of differences, the most conspicuous of which is the virtual absence of myosin in the spicule preparations. It is proposed that a lack of myosin produces a different cytoskeletal organization in the spicules. This enables phosphatidylserine to become exposed at the outer surface of the spicule membrane.

Evidence for the preferential interaction of glycophorin with negatively charged phospholipids.

Glycophorin, extracted from the erythrocyte membrane after treatment with lithium-diiodo-salicylate, still contains a significant amount of phospholipid, consisting predominantly of phosphatidylserine. Methods are described which lead to a full delipidation of the protein. After treatment with neuraminidase, delipidated glycophorin shows a preferential interaction with monolayers of negatively-charged phospholipids. This lipid-protein interaction is decreased by the presence of cholesterol in the lipid film.

Relation between various phospholipase actions on human red cell membranes and the interfacial phospholipid pressure in monolayers.

The action of purified phospholipases on monomolecular films of various interfacial pressures is compared with the action on erythrocyte membranes. The phospholipases which cannot hyorolyse phospholipids of the intact erythrocyte membrane, phospholipase C from Bacillus cereus, phospholipase A2 from pig pancreas and Crotalus adamanteus and phospholipase D from cabbage, can hydrolyse phospholipid monolayers at pressure below 31 dynes/cm only. The phospholipases which can hydrolyse phospholipids of the intact erythrocyte membrane, phospholipase C from Clostridium welchii phospholipase A2 from Naja naja and bee venom and sphingomyelinase from Staphylococcus aureus, can hydrolyse phospholipid monolayers at pressure above 31 dynes/cm. It is concluded that the lipid packing in the outer monolayer of the erythrocyte membrane is comparable with a lateral surface pressure between 31 and 34.8 dynes/cm.

Increased aminophospholipid translocase activity in human platelets during secretion.

Fluorescent labeled analogs of phosphatidylcholine (NBD-PC) and phosphatidylserine (NBD-PS) were used to study transport of phospholipids from the outer to the inner leaflet of the plasma membrane of human platelets. Platelets were stimulated with thrombin or Ca2(+)-ionophore at various extracellular [Ca2+]. No significant transport of NBD-PC could be observed either in resting or stimulated platelets. NBD-PS transport in platelets stimulated with thrombin (with or without extracellular Ca2+), or ionophore in the presence of EGTA, was enhanced 4-fold (t1/2 approximately 2 min) compared to unstimulated controls (t1/2 approximately 8 min). Stimulation with ionophore at extracellular [Ca2+] exceeding 8 microM caused a gradual decrease in inward transport of NBD-PS. At 100 microM Ca2+, NBD-PS transport becomes as slow as that of NBD-PC. We conclude that platelet activation by agonists that induce secretion without appreciable shedding is accompanied by an increase in translocase activity that maintains asymmetry during fusion which occurs during exocytosis.

Membrane fatty acid composition and endothelial cell functional properties.

In order to study the influence of endothelial cell fatty acid composition on various membrane related parameters, several in vitro methods were developed for manipulating the fatty acid content of human endothelial cell membranes. Changes in membrane fatty acid profile were induced by using fatty acid modified lipoproteins or free fatty acids. The largest changes in endothelial fatty acid composition were obtained by culturing the cells in media supplemented with specific free fatty acids. An increase in arachidonic acid content of endothelial phospholipids was induced by supplementation with saturated fatty acids or with arachidonic acid itself. A decrease in arachidonic acid content was obtained by supplementation with other unsaturated fatty acids. Under the experimental conditions used endothelial cells showed a low desaturase activity and a high elongase activity. Considerable alterations in membrane fatty acid composition did not greatly influence certain membrane related parameters such as polymorphonuclear leukocyte adherence and endothelial cell procoagulant activity. In general, for fatty acid modified endothelial cells an association between endogenous arachidonic acid content and total production of eicosanoids was found. This study demonstrates that considerable changes in membrane fatty acid profile affect endothelial cell arachidonic acid metabolism, but it also illustrates homeostasis at the level of endothelial cell functional activity.

Loss of membrane phospholipid asymmetry in platelets and red cells may be associated with calcium-induced shedding of plasma membrane and inhibition of aminophospholipid translocase.

Influx of calcium in platelets and red cells produces formation of vesicles shed from the plasma membrane. The time course of the shedding process closely correlates with the ability of both cells to stimulate prothrombinase activity when used as a source of phospholipid in the prothrombinase assay. This reflects increased surface exposure of phosphatidylserine, presumably resulting from a loss in membrane asymmetry. Evidence is presented that the shed vesicles have a random phospholipid distribution, while the remnant cells show a progressive loss of membrane phospholipid asymmetry when more shedding occurs. Removal of intracellular calcium produces a decrease of procoagulant activity of the remnant cells but not of that of the shed vesicles. This is consistent with reactivation of aminophospholipid translocase activity, being first inhibited by intracellular calcium and subsequently reactivated upon calcium removal. Involvement of aminophospholipid translocase is further supported by the observation that reversibility of procoagulant activity is also dependent on metabolic ATP and reduced sulfhydryl groups. The finding that this reversibility process is not apparent in shed vesicles may be ascribed to the absence of translocase or to a lack of ATP. These data support and extend the suggestion made by Sims et al. [1989) J. Biol. Chem. 264, 17049-17057) that membrane fusion, which is required for shedding to occur, produces transient flip-flop sites for membrane phospholipids. Furthermore, the present results indicate that scrambling of membrane phospholipids can only occur provided that aminophospholipid translocase is inactive.

Studies on sickled erythrocytes provide evidence that the asymmetric distribution of phosphatidylserine in the red cell membrane is maintained by both ATP-dependent translocation and interaction with membrane skeletal proteins.

In order to study factors which are involved in maintenance of phosphatidylserine (PS) asymmetry within the human red cell membrane, we measured the effect of ATP-depletion and of membrane skeleton/lipid bilayer uncoupling induced by sickling on the distribution of PS within the membrane bilayer of sickle cells. Trace amounts of radiolabeled PS were introduced into the outer membrane leaflet of both fresh and ATP-depleted reversibly sickled cells (RSCs), using a non-specific lipid transfer protein purified from bovine liver. The equilibration of the newly introduced PS over the two halves of the bilayer was monitored by treatment of the cells with phospholipase A2 which selectively hydrolyzes only those molecules present in the outer membrane leaflet. Within 1 h after insertion into fresh RSCs, only 10% of the labeled PS was accessible to the action of phospholipase A2. This fraction was markedly increased when the cells were subsequently deoxygenated. Prolonged deoxygenation of RSCs, deprived of their ATP after incorporation of radiolabeled PS, caused enhanced phospholipase A2-induced hydrolysis of radiolabeled PS. Similarly, phospholipase A2-induced hydrolysis of endogenous PS in intact RSCs was markedly enhanced when ATP-depleted, but not when fresh cells, were incubated under nitrogen for 3.5 h. Deoxygenated ATP-depleted RSCs markedly enhanced the rate of thrombin formation in the presence of purified coagulation factors Xa, Va, prothrombin and Ca2+. This enhancement appeared to be dependent on the duration of incubation under nitrogen. This phenomenon, indicating the presence of increasing amounts of endogenous PS in the outer membrane leaflet, was not observed when either fresh RSCs or ATP-depleted normal erythrocytes were incubated under nitrogen. Our present observations provide evidence that, in addition to the interaction of PS with the skeletal proteins, an ATP-dependent translocation of PS is required to maintain its absolute asymmetric distribution in the human erythrocyte membrane.