Importance of phosphatidylethanolamine for association of protein kinase C and other cytoplasmic proteins with membranes.

Biological membranes exhibit an asymmetric distribution of phospholipids. Phosphatidylserine (PS) is an acidic phospholipid that is found almost entirely on the interior of the cell where it is important for interaction with many cellular components. A less well understood phenomenon is the asymmetry of the neutral phospholipids, where phosphatidylcholine (PC) is located primarily on exterior membranes while phosphatidylethanolamine (PE) is located primarily on interior membranes. The effect of these neutral phospholipids on protein-phospholipid associations was examined using four cytoplasmic proteins that bind to membranes in a calcium-dependent manner. With membranes containing PS at a charge density characteristic of cytosolic membranes, protein kinase C and three other proteins with molecular masses of 64, 32, and 22 kDa all showed great selectively for membranes containing PE rather than PC as the neutral phospholipid; the calcium requirements for membrane-protein association of the 64- and 32-kDa proteins were about 10-fold lower with membranes containing PE; binding of the 22-kDa protein to membranes required the presence of PE and could not even be detected with membranes containing PC. Variation of the PS/PE ratio showed that membranes containing about 20% PS/60% PE provided optimum conditions for binding and were as effective as membranes composed of 100% PS. Thus, PE, as a phospholipid matrix, eliminated the need for membranes with high charge density and/or reduced the calcium concentrations needed for protein-membrane association. A surprising result was that PKC and the 64- and 32-kDa proteins were capable of binding to neutral membranes composed entirely of PE/PC or PC only. The different phospholipid headgroups altered only the calcium required for membrane-protein association. For example, calcium concentrations at the midpoint for association of the 64-kDa protein with membranes containing PS, PE/PC, or PC occurred at 6, 100, and 20,000 microM, respectively. Thus, biological probes detected major differences in the surface properties of membranes containing PE versus PC, despite the fact that both of these neutral phospholipids are often thought to provide "inert" matrices for the acidic phospholipids. The selectivity for membranes containing PE could be a general phenomenon that is applicable to many cytoplasmic proteins. The present study suggested that the strategic location of PE on the interior of the membranes may be necessary to allow some membrane-protein associations to occur at physiological levels of calcium and PS.     

Purification, characterization and substrate specificity of rabbit lung phospholipid transfer proteins.

Three phospholipid transfer proteins, namely proteins I, II and III, were purified from the rabbit lung cytosolic fraction. The molecular masses of phospholipid transfer proteins I, II and III are 32 kilodaltons (kDa), 22 kDa and 32 kDa, respectively; their isoelectric point values are 6.5, 7.0 and 6.8, respectively. Phospholipid transfer proteins I and III transferred phosphatidylcholine (PC) and phosphatidylinositol (PI) from donor unilamellar liposomes to acceptor multilamellar liposomes; protein II transferred PC but not PI. All the three phospholipid transfer proteins transferred phosphatidylethanolamine poorly and showed no tendency to transfer triolein. The transfer of [14C]PC from unilamellar liposomes to multilamellar liposomes facilitated by each protein was affected differently by the presence of acidic phospholipids in the PC unilamellar liposomes. In an equal molar ratio of acidic phospholipid and PC, phosphatidylglycerol (PG) reduced the activities of proteins I and III by 70% (P = 0.0004 and 0.0032, respectively) whereas PI and phosphatidylserine (PS) had an insignificant effect. In contrast, the protein II activity was stimulated 2-3-times more by either PG (P = 0.0024), PI (P = 0.0006) or PS (P = 0.0038). In addition, protein II transferred dioleoylPC (DOPC) about 2-times more effectively than dipalmitoylPC (DPPC) (P = 0.0002), whereas proteins I and III transferred DPPC 20-40% more effectively than DOPC but this was statistically insignificant. The markedly different substrate specificities of the three lung phospholipid transfer proteins suggest that these proteins may play an important role in sorting intracellular membrane phospholipids, possibly including lung surfactant phospholipids.     

Binding of bovine factor Va to phosphatidylcholine membranes.

The interaction of bovine factor Va with phosphatidylcholine membranes was examined using four different fluorescence techniques: 1) changes in the fluorescence anisotropy of the fluorescent membrane probe 1,6-diphenyl-1,3,5-hexatriene (DPH) to monitor the interaction of factor Va with 1,2-dimyristoyl-3-sn-phosphatidylcholine (DMPC) small unilamellar vesicles (SUVs), 2) changes in the fluorescence anisotropy of N-(lissamine rhodamine B sulfonyl) diacyl phosphati-dylethanolamine (Rh-PE) incorporated into SUVs prepared from 1-palmitoyl-2-oleoyl-3-sn-phosphatidylcholine (POPC), 3) changes in the fluorescence anisotropy of fluorescein-labeled factor Va (labeled in the heavy chain) upon interaction with POPC SUVs, 4) fluorescence energy transfer from fluorescein-labeled factor Va to rhodamine-labeled POPC SUVs. In the first two sets of experiments, labeled lipid vesicles were titrated with unlabeled protein, whereas, in the latter two types of experiments, labeled factor Va was titrated with vesicles. For the weak binding observed here, it was impossible from any one binding experiment to obtain precise estimates of the three parameters involved in modeling the lipid-protein interaction, namely, the dissociation constant Kd, the stoichiometry of binding i, and the saturation value of the observable Rmax from any one experiment. However, a global analysis of the four data sets involving POPC SUVs yielded a stable estimate of the binding parameters (Kd of approximately 3.0 microM and a stoichiometry of approximately 200 lipids per bound factor Va). Binding to DMPC SUVs may be of slightly higher affinity. These observations support the contention that association of factor Va with a membrane involves a significant acidic-lipid-independent interaction along with the more commonly accepted acidic-lipid-dependent component of the total binding free energy.     

Association of a protein with membrane vesicles at the collisional limit: studies with blood coagulation factor Va light chain also suggest major differences between small and large unilamellar vesicles

Vesicle size can be a very sensitive modulator of protein-membrane association. In addition, reactions at the collisional limit may be characteristic of many types of protein-membrane or protein-receptor interactions. To probe these effects quantitatively, we analyzed the association of blood clotting factor Va light chain (Va-LC) with phospholipid vesicles of 15-150-nm radius. The number of protein binding sites per vesicle was approximately proportional to vesicle surface area. Association rates approached the collisional limit, and the activation energy for the association reaction was 4.5 +/- 0.5 kcal/mol. In agreement with diffusional theory for this type of interaction at the collisional limit, the observed association rate constant for filling all sites was approximately proportional to the inverse of vesicle radius. This general property has important implications for many systems such as blood coagulation including possible slower association rates and higher Km values for reactions involving whole cells relative to those obtained for phospholipid vesicles. Dissociation rate constants for reactions that are near the collisional limit should also be proportional to the inverse of vesicle size if diffusional parameters are the only factors influencing dissociation. However, Va-LC bound to small unilamellar vesicles (SUVs, less than or equal to 15-nm radius) gave slower dissociation rates than Va-LC bound to large unilamellar vesicles (LUVs, greater than or equal to 35-nm radius). This indicated a change in KI, the intrinsic protein-phospholipid affinity constant for LUVs vs SUVs. The cumulative effect of association and dissociation rates resulted in higher affinity of Va-LC for SUVs than LUVs under equilibrium conditions. The latter was corroborated by competition binding studies. Furthermore, the temperature dependence of both rate constants indicated an entirely entropy-driven binding to LUVs but a largely enthalpy-driven binding to SUVs. Interactions which are largely entropic are thought to be ionic in nature. The differences observed between binding to LUVs and SUVs may reflect thermodynamic differences between these types of phospholipid structures.     

In vitro lipid transfer assays of phosphatidylinositol transfer proteins provide insight into the in vivo mechanism of ligand transfer

Fluorescence resonance energy transfer (FRET) assays and membrane binding determinations were performed using three phosphatidylinositol transfer proteins, including the yeast Sec14 and two mammalian proteins PITPα and PITPβ. These proteins were able to specifically bind the fluorescent phosphatidylcholine analogue NBD-PC ((2-(12-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino)dodecanoyl-1-hexadecanoyl-sn-glycero-3-phosphocholine)) and to transfer it to small unilamellar vesicles (SUVs). Rate constants for transfer to vesicles comprising 100% PC were slower for all proteins than when increasing percentages of phosphatidylinositol were incorporated into the same SUVs. The rates of ligand transfer by Sec14 were insensitive to the inclusion of equimolar amounts of another anionic phospholipid phosphatidylserine (PS), but the rates of ligand transfer by both mammalian PITPs were strikingly enhanced by the inclusion of phosphatidic acid (PA) in the receptor SUV. Binding of Sec14 to immobilized bilayers was substantial, while that of PITPα and PITPβ was 3–7 times weaker than Sec14 depending on phospholipid composition. When small proportions of the phosphoinositide PI(4)P were included in receptor SUVs (either with PI or not), Sec14 showed substantially increased rates of NBD-PC pick-up, whereas the PITPs were unaffected. The data are supportive of a role for PITPβ as functional PI transfer protein in vivo, but that Sec14 likely has a more elaborate function.     

Mutation of Serine 41 in the Neuron-Specific Protein B-50 (GAP-43) Prohibits Phosphorylation by Protein Kinase C

The neuron-specific, calmodulin-binding protein B-50 (also known as GAP-43, F1, or neuromodulin) is an endogenous substrate of protein kinase C (PKC). PKC exclusively phosphorylates Ser residues in B-50. As potential phosphorylation sites for PKC, Ser41, Ser110, and Ser122 were indicated, of which Ser41 is contained in the sequence ASF, which matches with the sequence of a synthetic PKC substrate. N-terminally 35S-labeled B-50, produced from cDNA, was subjected to digestion with Staphylococcus aureus V8 protease (SAP). Consecutively, 35S-labeled 28- and 15-kDa fragments were formed, similar to those after digestion of 32P-labeled B-50. In a previous study, we showed that the 32P-labeled 15-kDa SAP fragment contains all 32P radioactivity. The present data indicate that it contains the N-terminus of B-50 as well. The 15-kDa fragment, with a calculated length ranging from amino acid residue 1 to 65, contains only one potential PKC phosphorylation site, at Ser41. Mutagenesis of Ser41 into Thr or Ala resulted in recombinant B-50 products with mobilities on two-dimensional electrophoresis similar to those of the nonmutated recombinant B-50 and the rat brain B-50. Only [Ser41]B-50 was phosphorylated by PKC, whereas [Thr41]- or [Ala41]B-50 did not show any phosphorylation at the positions indicated on the immunoblots. This leads us to the conclusion that Ser41 is the sole phosphorylation site for PKC in vitro.     

Interaction of blood coagulation factor Va with phospholipid vesicles examined by using lipophilic photoreagents

Two different lipophilic photoreagents, [3H]adamantane diazirine and 3-(trifluoromethyl)-3-(m-[125I]iodophenyl)diazirine (TID), have been utilized to examine the interactions of blood coagulation factor Va with calcium, prothrombin, factor Xa, and, in particular, phospholipid vesicles. With each of these structurally dissimilar reagents, the extent of photolabeling of factor Va was greater when the protein was bound to a membrane surface than when it was free in solution. Specifically, the covalent photoreaction with Vl, the smaller subunit of factor Va, was 2-fold higher in the presence of phosphatidylcholine/phosphatidylserine (PC/PS, 3:1) vesicles, to which factor Va binds, than in the presence of 100% PC vesicles, to which the protein does not bind. However, the magnitude of the PC/PS-dependent photolabeling was much less than has been observed previously with integral membrane proteins. It therefore appears that the binding of factor Va to the membrane surface exposes Vl to the lipid core of the bilayer, but that only a small portion of the Vl polypeptide is exposed to, or embedded in, the bilayer core. Addition of either prothrombin or active-site-blocked factor Xa to PC/PS-bound factor Va had little effect on the photolabeling of Vl with TID, but reduced substantially the covalent labeling of Vh, the larger subunit of factor Va. This indicates that prothrombin and factor Xa each cover nonpolar surfaces on Vh when the macromolecules associate on the PC/PS surface. It therefore seems likely that the formation of the prothrombinase complex involves a direct interaction between Vh and factor Xa and between Vh and prothrombin.(ABSTRACT TRUNCATED AT 250 WORDS)     

Purification and partial characterization of lysosomal neuraminidase from human placenta

Lysosomal neuraminidase and beta-galactosidase are present in a complex together with a 32-kDa protective protein. This complex has been purified and the different components have been dissociated using potassium isothiocyanate (KSCN) treatment. beta-Galactosidase remains catalytically active, but neuraminidase loses its activity upon dissociation. The inactive dissociated neuraminidase was purified by removing the remaining non-dissociated beta-galactosidase/protective protein complex using beta-galactosidase-specific affinity chromatography. The dissociated neuraminidase material shows two major polypeptides on SDS-PAGE with an apparent molecular mass of 76 kDa and 66 kDa. Subsequently the 32-kDa protective protein was dissociated from the beta-galactosidase/protective protein complex, and purified. Antibodies raised against the dissociated inactive neuraminidase preparation specifically immunoprecipitate the active neuraminidase present in the complex with beta-galactosidase and protective protein. By immunoblotting evidence is provided that the 76-kDa protein is a subunit of neuraminidase which, in association with the 32-kDa protective protein, is essential for neuraminidase activity.     

Binding of microtubule-associated protein 2 and tau to the intermediate filament reassembled from neurofilament 70-kDa subunit protein. Its regulation by calmodulin

Two major brain microtubule-associated proteins (MAPs), MAP2 and tau, were found to bind to the intermediate filaments reassembled from neurofilament 70-kDa subunit protein (= 70-kDa filaments). The binding was saturable. The apparent dissociation constant (KD) for the binding of MAP2 to the 70-kDa filaments was estimated to be 4.8 X 10(-7) M, and the maximum binding reached 1 mol of MAP2/approximately 30 mol of 70-kDa protein. The apparent KD for the tau binding was 1.6 X 10(-6) M, and the maximum binding was 1 mol of tau/approximately 3 mol of 70-kDa protein. It was also found that MAP2 and tau did not compete with each other for binding to the 70-kDa filaments. Most interestingly, calmodulin, a ubiquitous Ca2+-binding protein in eukaryotic cells, was found to inhibit the binding of MAP2 and tau to the 70-kDa filaments. The inhibition by calmodulin was regulated by changes in Ca2+ concentration around 10(-6) M, and was canceled by trifluoperazine, a calmodulin inhibitor.     

Small and large unilamellar vesicle membranes as model system for bile acid diffusion in hepatocytes.

Uptake of bile acids into the liver cell occurs via active transport or passive diffusion. In a model system, passive diffusion was studied in liposomes using pyranine fluorescence. Rate constants for the diffusion of diverse more polar or more apolar bile acids were examined. Hydrophobic lithocholic acid (LCA) revealed a maximal rate constant of 0.057 s(-1); with the polar ursodeoxycholic acid (UDCA), the value was 0.019 s(-1). UDCA (3 mol%) effectively decreased the rate constant of 0.1 mM chenodeoxycholic acid (CDCA), whereas cholesterol reached a similar decrease only between 5 and 10 mol%. At higher concentrations of CDCA (above 1 mM) or LCA (0.3-0.4 mM), breaking up of liposomal structure was confirmed by light-scattering decrease and increase of carboxyfluorescein fluorescence. Changes in lipid composition of phosphatidylcholine (PC)- small unilamellar vesicles (SUVs) or large unilamellar vesicles (LUVs) also caused decreasing rate constants. For a cardiolipin (CL):PC ratio of 1:20 the CDCA (0.1 mM) rate constant was 71% lower (0.015 s(-1)) and for a sphingomyelin (SM):PC ratio of 2:1 the rate constant was 50% lower (0.026 s(-1)). Changes in membrane fluidity were detected using membrane anisotropy measurements with the 1,6-diphenyl-1,3, 5-hexatriene (DPH) method. Membrane fluidity was reduced with cholesterol- but not with CL- or SM-containing SUVs (ratio: cholesterol, CL, SM:PC of 1:5). This model system is currently used for the analysis of more complex lipid vesicles resembling the plasma/hepatocyte membrane, which is either stabilized or destabilized by appropriate conditions. The results should become clinically relevant.     

Active-site copper and zinc ions modulate the quaternary structure of prokaryotic Cu,Zn superoxide dismutase

The influence of the constitutive metal ions on the equilibrium properties of dimeric Photobacterium leiognathi Cu,Zn superoxide dismutase has been studied for the wild-type and for two mutant protein forms bearing a negative charge in the amino acid clusters at the dimer association interface. Depletion of copper and zinc dissociates the two mutant proteins into monomers, which reassemble toward the dimeric state upon addition of stoichiometric amounts of zinc. Pressure-dependent dissociation is observed for the copper-depleted wild-type and mutated enzymes, as monitored by the fluorescence shift of a unique tryptophan residue located at the subunit association interface. The spectral shift occurs slowly, reaching a plateau after 15-20 minutes, and is fully reversible. The recovery of the original fluorescence properties, after decompression, is fast (less than four minutes), suggesting that the isolated subunit has a relatively stable structure, and excluding the presence of stable intermediates during the dimer-monomer transition. The dimer dissociation process is still incomplete at 6.5 kbar for the copper-depleted wild-type and mutated enzymes, at variance with what is generally observed for oligomeric proteins that dissociate below 3 kbar. Measurement of the degree of dissociation, at two different protein concentrations, allows us to calculate the standard volume variation upon association, Delta V, and the dissociation constant K(d0), at atmospheric pressure, (25 ml/mol and 3 x 10(-7)M, respectively). The holoprotein is fully dimeric even at 6.5 kbar, which allows us to evaluate a lower Delta G degrees limit of 11.5 kcal/mol, corresponding to a dissociation constant K(d0)<10(-9)M.     

Preferential lipid association and mode of penetration of apocytochrome c in mixed model membranes as monitored by tryptophanyl fluorescence quenching using brominated phospholipids

The fluorescence of the single tryptophan residue at position 59 in apocytochrome c, the biosynthetic precursor of the inner mitochondrial membrane protein cytochrome c, was studied in small unilamellar vesicles composed of phosphatidylserine (PS) and phosphatidylcholine (PC) with or without specifically Br-labelled acyl chains at the sn-2 position. The protein has a very high affinity for PS-containing vesicles (dissociation constant Kd less than 1 microM). From the relative quenching efficiency by the brominated phospholipids, it could be concluded that the protein specifically associates with the PS component in mixed vesicles and that maximal quenching occurred with phospholipids in which the bromine was present at the 6,7-position of the 2-acyl chain suggesting that (part of) the bound protein penetrates 7-8 A deep into the hydrophobic core of the bilayer.     

Lipocortins are major substrates for protein kinase C in extracts of human neutrophils.

We have identified two major proteins in human neutrophils that are phosphorylated in vitro by protein kinase C (PKC) as lipocortins III and a fragment of a lipocortin-like 68-kDa protein. In electroporated cells, the 68-kDa protein was phosphorylated during stimulation of the cells with either FMLP or PMA. Lipocortins are of interest because of their Ca2(+)- and phospholipid-dependent actin binding properties and ability to inhibit phospholipase A2. Two crude fractions of enzymes and proteins exposed to [gamma-32]PATP in the presence of Ca2+, Mg2+, phosphatidylserine and 1,2-oleoyl-acetyl-rac-glycerol were analyzed by gel electrophoresis and autoradiography. A number of proteins in a detergent-free fraction, including proteins at 36 and 32 kDa, were phosphorylated in the presence of these cofactors. In contrast, only two major proteins (35 and 32 kDa) were phosphorylated in a detergent-extracted fraction. Phosphorylation of the 36, 35, and 32 kDa proteins required the presence of Ca2+, Mg2+, and phosphatidylserine in our soluble fraction and detergent extract, indicating PKC-dependent phosphorylation. The 32-kDa protein phosphorylated in both the soluble fraction and detergent extract was identified as lipocortin III by immunoprecipitation with a cross-reactive antibody that recognized lipocortin I and comparison of cyanogen bromide (CNBr) cleavage patterns of this protein with a lipocortin III standard. The 68-kDa protein was identified as a lipocortin VI-like protein by immunoprecipitation with anti-calelectrin. Additionally, the CNBr cleavage pattern of the 68-kDa protein was similar to that of the 36-kDa protein phosphorylated in our soluble fraction. Autoradiograms of the 68- and 36-kDa fragments immunoprecipitated from our soluble fraction with anticalelectrin and cleaved with CNBr showed that both of these proteins were phosphorylated in this sample. Because phosphorylation is known to change the functional characteristics of the lipocortins, the potential exists to link PKC and lipocortins in neutrophils to regulation of granulemembrane interactions or mediation of inflammation.     

Purification of unactivated progesterone receptor and identification of novel receptor-associated proteins

Progesterone receptor complexes were purified from crude cytosol in a rapid, gentle, one-step procedure using anti-receptor monoclonal antibodies covalently attached to an agarose resin. Five nonreceptor proteins specifically co-purified with unactivated avian progesterone receptor; these proteins had molecular masses of approximately 90, 70, 54, 50, and 23 kDa. The 90- and 70-kDa proteins have been previously identified as the 90-kDa heat shock protein and a member of the 70-kDa heat shock protein family, respectively. The 54-, 50-, and 23-kDa proteins have not been previously described as associated with avian progesterone receptor. Two-dimensional gel electrophoresis revealed charge heterogeneities for all five proteins. Except for p70, each could be dissociated from receptor by salt, a process inhibited by sodium molybdate. However, molybdate was not required for protein association with receptor in low ionic strength. Following progesterone treatment in vivo p70 still co-purified with cytosolic receptor although the other affiliated proteins were reduced, suggesting hormone-dependent dissociation in conjunction with receptor activation. One of the proteins, p54, displayed in vitro hormone-dependent dissociation which was not prevented by molybdate.     

Phosphatidylcholine could be the source of 1,2-DAG which activates protein kinase C in EGF-stimulated colon carcinoma cells (HT29)

In our previous study (A. Balogh et al, Cell. Signalling 5 (6), 795-802, 1993.), we have shown that epidermal growth factor (EGF) increased protein kinase C (PKC) activities in colon carcinoma cell line (HT29), possibly through the increased 1,2-diacylglycerol (1,2-DAG) production via phosphatidylcholine (PC). Here we investigate the effect of the well-known PKC activator 12-O-tetradecanoyl-2 phorbol-13-acetate (TPA), on the levels of ³²P incorporation into EGF induced phosphatidylinositols (PI, PI4P, PI4, 5P₂) and different phospholipids (PC, PA, PS) as well as on induced tyrosine kinase activity. TPA significantly decreased the effects of EGF and it had the biggest inhibitory effect on EGF induced PC level. These data support our contention that PC plays an important role in the activation of PKC via 1,2-DAG production in the EGF stimulated pathway.     

Association of protein kinase C with phospholipid vesicles

The Ca2+- and phospholipid-dependent protein kinase, protein kinase C (PKC), was purified from bovine brain by a modified procedure that provided sufficient quantities of stable protein for analysis of physical properties of protein-membrane binding. The binding of PKC to phospholipid vesicles of various compositions was investigated by light-scattering and fluorescence energy transfer measurements. The binding properties for membranes of low phosphatidylserine (PS) content were consistent with a peripheral membrane association; PKC showed Ca2+ -dependent binding to phospholipid vesicles containing phosphatidylserine, phosphatidylinositol, or phosphatidylglycerol. Membranes containing 0-20% PS (the remainder of the phospholipid was phosphatidylcholine) bound less protein than membranes containing greater than 20% PS; the factor limiting protein binding to membranes containing low PS appeared to be the availability of acidic phospholipids. Increasing the PS content above 20% did not increase the amount of membrane-bound protein at saturation, and the limiting factor was probably steric packing of protein on the membrane surface. The membranes bound about 1 g of protein/g of phospholipid at steric saturation. Binding was of relatively high affinity (Kd less than 5 nM), and the association rate was rapid on the time scale of the experiments. Addition of ethylene glycol bis(beta-aminoethyl ether)-N,N,N',N'-tetraacetic acid to phospholipid-bound PKC caused dissociation of the complex, and the properties of this dissociation indicated an equilibrium binding of protein to membrane. However, only partial dissociation of PKC was achieved when the PS content of the vesicles exceeded 20%. A number of comparisons revealed that binding of protein to the membrane, even in the presence of phorbol esters, was insufficient for development of enzyme activity.(ABSTRACT TRUNCATED AT 250 WORDS)     

Nerve Growth Factor Receptors in Human Neuroblastoma Cells

Receptors for the nerve growth factor protein (NGFR) present in the human neuroblastoma cell line LAN-1 were characterized. LAN-1 cells display high-affinity (type I, with KD value of 5.9 X 10(-11) M) and low-affinity (type II, with KD value of 9.2 X 10(-9) M) binding to NGF. NGFR were fractionated by preparative isoelectric focusing in a granulated gel (PEGG). High-affinity binding was found in the 5.9-6.2 pH region of the PEGG, and low-affinity binding in the 4.6-4.8 and 8.8-9.3 pH ranges. After further analysis by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) we observed both 92.5- and 200-kDa molecular species associated with NGF binding activity. The 200-kDa protein was found in fractions displaying high-affinity NGF binding and the 92.5-kDa protein in fractions displaying low-affinity NGF binding. Equilibrium binding analysis of NGF in PEGG fractions confirmed the presence of two specific saturable binding sites with KD values similar to those observed for whole dissociated cells. When NGFR II activity from the acidic region of the PEGG chromatogram was incubated with NGFR II from the basic region of the PEGG chromatogram, there was no change in NGF binding or in the number of apparent NGF receptors. However, incubation of these same fractions with a fraction having only NGFR I showed an apparent increase in high-affinity NGF binding and a decrease in low-affinity NGF binding. Immunoprecipitation of this "mixed" fraction and analysis on SDS-PAGE under reduced and nonreduced conditions showed 200-kDa and 92.5-kDa proteins under nonreduced conditions and a 92.5-kDa protein under reduced conditions. Our findings are consistent with the hypothesis that there are two distinct NGF receptors in NGF-responsive cells. The interconvertibility of low- and high-affinity receptors and the possible existence of a modulator type protein or of "silent" type receptors are also in agreement with our findings.     

A domain of membrane-bound blood coagulation factor Va is located far from the phospholipid surface. A fluorescence energy transfer measurement

The larger subunit of blood coagulation factor Va was covalently labeled with iodoacetamido derivatives of fluorescein and rhodamine without loss of functional activity, as measured by either the one-stage clotting assay or the ability to accelerate prothrombin activation in a purified system. The spectral properties of the dyes were not altered by the presence or absence of the smaller subunit of factor Va, Ca2+, prothrombin, factor Xa, or phosphatidylcholine/phosphatidylserine (PC/PS, 4:1) vesicles. When fluorescein-labeled protein (factor VaF) was titrated with PC/PS vesicles containing either octadecylrhodamine or 5-(N-hexadecanoylamino)eosin, fluorescence energy transfer was observed between the protein-bound donor dyes and the acceptor dyes at the outer surface of the phospholipid bilayer. The extent of energy transfer correlated directly with the extent of protein binding to the vesicles monitored by light scattering. The distance of closest approach between the fluorescein on factor Va and the bilayer surface averaged 90 A for the two different acceptors. Association of factor VaF with factor Xa on the phospholipid surface reduced this separation by 7 A, but association with prothrombin did not alter the distance between the labeled domain on factor VaF and the surface. The efficiency of diffusion-enhanced energy transfer between rhodamine-labeled factor Va and terbium dipicolinate entrapped inside PC/PS vesicles was less than 0.01, consistent with the location of the dye far above the inner surface of the vesicle. Thus, a domain of membrane-bound factor Va is located a minimum of 90 A above the phospholipid surface.(ABSTRACT TRUNCATED AT 250 WORDS)     

The effect of membrane composition on the hemostatic balance

The phospholipid composition requirements for optimal prothrombin activation and factor Va inactivation by activated protein C (APC) anticoagulant were examined. Vesicles composed of phosphatidylethanolamine (PE) and phosphatidylcholine (PC) supported factor Va inactivation relatively well. However, optimal factor Va inactivation still required relatively high concentrations of phosphatidylserine (PS). In addition, at a fixed concentration of phospholipid, PS, and APC, vesicles devoid of PE never attained a rate of factor Va inactivation achievable with vesicles containing PE. Polyunsaturation of any vesicle component also contributed significantly to APC inactivation of factor Va. Thus, PE makes an important contribution to factor Va inactivation that cannot be mimicked by PS. In the absence of polyunsaturation in the other membrane constituents, this contribution was dependent upon the presence of both the PE headgroup per se and unsaturation of the 1,2 fatty acids. Although PE did not affect prothrombin activation rates at optimal PS concentrations, PE reduced the requirement for PS approximately 10-fold. The Km(app) for prothrombin and the Kd(app) for factor Xa-factor Va decreased as a function of increasing PS concentration, reaching optimal values at 10-15% PS in the absence of PE but only 1% PS in the presence of PE. Fatty acid polyunsaturation had minimal effects. A lupus anticoagulant immunoglobulin was more inhibitory to both prothrombinase and factor Va inactivation in the presence of PE. The degree of inhibition of APC was significantly greater and much more dependent on the phospholipid composition than that of prothrombinase. Thus, subtle changes in the phospholipid composition of cells may control procoagulant and anticoagulant reactions differentially under both normal and pathological conditions.     

The interaction of protein kinase C and other specific cytoplasmic proteins with phospholipid bilayers

The role of lipid composition in the interaction of purified protein kinase C with large unilamellar vesicles was determined by the extent of photolabelling of the enzyme with 5-[125I]iodonaphthalene-I-azide. The protein kinase C was only slightly labelled when exposed to phosphatidylcholine (PC) liposomes. The addition of phorbol 12-myristate 13-acetate (PMA) or of diacylglycerol to the PC liposomes enhanced significantly the labelling of the protein kinase C at low calcium concentrations. A further enhancement in the photolabelling of the protein kinase C was observed in liposomes containing 2% phosphatidylserine (PS). At low calcium concentrations, the binding of the enzyme to these liposomes increased in the presence of added PMA or diacylglycerol. Raising the levels of PS beyond 2% in the liposomes did not enhance the binding of the protein kinase C. However, when the enzymatic activity of the protein kinase C was measured using basic histones as substrates, maximum phosphorylation was obtained in liposomes with a PC to PS ratio of 1. The fact that the translocation of the protein kinase C from solution to the surface of the liposomes could be monitored by its labelling with 5-iodonaphthalene 1-azide prompted us to determine whether other cytoplasmic proteins might share this property. The interaction of cytoplasmic proteins from HeLa cells with PC liposomes gave trace labelling irrespective of whether calcium was added. When the HeLa cell cytoplasmic proteins were allowed to interact with liposomes containing PS, selective 5-iodonaphthalene-1-azide photolabelling was observed in distinct proteins. Addition of calcium and of PMA or diacylglycerol modified the labelling of some but not all of these proteins. These results suggest that the methodology developed might serve to identify proteins that move to the membrane during stimulation of cells by phorbol esters or by growth factors which induce the generation of diacylglycerol. These results also suggest a role for the phospholipid composition of the plasma membrane (or any intracellular membrane) in the modulation of the activation processes of specific phospholipid-dependent proteins, in particular protein kinase C.