The effect of metal binding on the structure of annexin V and implications for membrane binding.

The structure of annexin V, crystallised in the presence of two calcium or barium ions for each protein molecule, was solved by molecular replacement to 0.24 nm resolution. The two metal ions are found in domains I and IV, i.e. on the same side of the channel that lies in the centre of the molecule. The structures of the barium and calcium form are extremely close, the only differences localised in the metal-binding sites that lie on the surface of the molecule. The occupancies of the metal ions, however, are lower for barium than for calcium, expressing the lower affinity of the protein for the former. The packing of the annexin molecules in the crystal asymmetric unit may represent a model for the calcium driven association of membrane-bound annexins that leads to membrane fusion.     

Phospholipid binding of annexin V: effects of calcium and membrane phosphatidylserine content.

We studied the binding of fluorescein-labeled annexin V (placental anticoagulant protein I) to small unilamellar phospholipid vesicles at 0.15 M ionic strength as a function of calcium concentration and membrane phosphatidylserine (PS) content. As the mole percentage of PS in the membrane increased from 10 to 50%, the stoichiometry of binding decreased hyperbolically from 1100 mol phospholipid/mol annexin V to a limiting value of 84 mol/mol for measurements made at 1.2 mM CaCl2. Over the same range of PS content, Kd remained approximately constant at 0.036 +/- 0.011 nM. A similar hyperbolic decrease in stoichiometry was observed with vesicles containing 10 or 20% PS when the calcium concentration was increased from 0.4 to 10 mM. Thus, the density of membrane binding sites is strongly dependent on the membrane PS content and calcium concentration. The effect of calcium on annexin V-membrane binding is proposed to be due to the formation of phospholipid-calcium complexes, to which the protein binds, rather than to an allosteric effect of calcium on protein-phospholipid affinity.     

Crystal and molecular structure of human annexin V after refinement. Implications for structure, membrane binding and ion channel formation of the annexin family of proteins.

Two crystal forms (P6(3) and R3) of human annexin V have been crystallographically refined at 2.3 A and 2.0 A resolution to R-values of 0.184 and 0.174, respectively, applying very tight stereochemical restraints with deviations from ideal geometry of 0.01 A and 2 degrees. The three independent molecules (2 in P6(3), 1 in R3) are similar, with deviations in C alpha positions of 0.6 A. The polypeptide chain of 320 amino acid residues is folded into a planar cyclic arrangement of four repeats. The repeats have similar structures of five alpha-helical segments wound into a right-handed compact superhelix. Three calcium ion sites in repeats I, II and IV and two lanthanum ion sites in repeat I have been found in the R3 crystals. They are located at the convex face of the molecule opposite the N terminus. Repeat III has a different conformation at this site and no calcium bound. The calcium sites are similar to the phospholipase A2 calcium-binding site, suggesting analogy also in phospholipid interaction. The center of the molecule is formed by a channel of polar charged residues, which also harbors a chain of ordered water molecules conserved in the different crystal forms. Comparison with amino acid sequences of other annexins shows a high degree of similarity between them. Long insertions are found only at the N termini. Most conserved are the residues forming the metal-binding sites and the polar channel. Annexins V and VII form voltage-gated calcium ion channels when bound to membranes in vitro. We suggest that annexins bind with their convex face to membranes, causing local disorder and permeability of the phospholipid bilayers. Annexins are Janus-faced proteins that face phospholipid and water and mediate calcium transport.     

Ca(2+)-dependent annexin self-association on membrane surfaces

Annexin self-association was studied with 90 degrees light scattering and resonance energy transfer between fluorescein (donor) and eosin (acceptor) labeled proteins. Synexin (annexin VII), p32 (annexin IV), and p67 (annexin VI) self-associated in a Ca(2+)-dependent manner in solution. However, this activity was quite labile and, especially for p32 and p67, was not consistently observed. When bound to chromaffin granule membranes, the three proteins consistently self-associated and did so at Ca2+ levels (pCa 5.0-4.5) approximately 10-fold lower than required when in solution. Phospholipid vesicles containing phosphatidylserine and phosphatidylethanolamine (1:1 or 1:3) were less effective at supporting annexin polymerization than were those containing phosphatidylserine and phosphatidylcholine (1:0, 1:1, or 1:3). The annexins bound chromaffin granule membranes in a positively cooperative manner under conditions where annexin self-association was observed, and both phenomena were inhibited by trifluoperazine. Ca(2+)-dependent chromaffin granule membrane aggregation, induced by p32 or synexin, was associated with intermembrane annexin polymerization at Ca2+ levels less than pCa 4, but not at higher Ca2+ concentrations, suggesting that annexin self-association may be necessary for membrane contact at low Ca2+ levels but not at higher Ca2+ levels where the protein may bind two membranes as a monomer.     

Site-specific mutagenesis of annexin V: role of residues from Arg-200 to Lys-207 in phospholipid binding.

Annexin V (placental anticoagulant protein I) binds tightly to anionic phospholipid vesicles in the presence of calcium. Four mutant proteins were expressed in Escherichia coli in which Ala replaced one of the following residues in the third repeat of annexin V: Arg-200, His-204, Arg-206, or Lys-207. In a competitive fluorescence quenching assay, the wild-type recombinant protein had the same affinity for phosphatidylserine-containing vesicles as the placentally derived protein. The affinity of the four mutant proteins for phosphatidylserine-containing vesicles was unchanged relative to wild-type protein. We conclude that His-204 and adjacent basic residues, including the highly conserved Arg-200 residue, are not required for high-affinity phospholipid binding.     

Calcium-dependent binding of annexin 12 to phospholipid bilayers: stoichiometry and implications

Annexins (ANXs) are a superfamily of proteins whose functional hallmark is Ca2+-dependent binding to anionic phospholipids. Their core domains are usually composed of a 4-fold repeat of a conserved amino acid sequence, with each repeat containing a type II Ca2+ binding site that is generally thought to mediate Ca2+-dependent binding to the membrane. We now report that ANX12 binding to phospholipid vesicles is highly cooperative with respect to Ca2+ concentration (Hill constant approximately 7), thereby suggesting that more than the four well-characterized type II Ca2+ binding sites are involved in phospholipid binding. Two independent approaches, a novel 45Ca2+ copelleting assay and isothermal titration calorimetry, indicate a stoichiometry of approximately 12 mol of Ca2+/mol of ANX12 for binding to phospholipid vesicles. On the basis of the "low-affinity" Ca2+-binding sites in a number of ANX X-ray crystal structures, we propose a model for ANX12 bilayer binding that involves three types of Ca2+ sites in each of the four repeats. In this model, there is a complementarity between the spacing of the ANX12 Ca2+ binding sites and the spacing of the phospholipid headgroups in bilayers. We tested the implications of the model by manipulating the physical state of vesicles composed of phospholipids with saturated acyl chains with temperature and measuring its influence on ANX12 binding. ANX12 bound to vesicles in a Ca2+-dependent manner when the vesicles were in the liquid crystal phase but not when the phospholipid was in the gel phase. Furthermore, ANX12 bound initially to fluid bilayers remained bound when cooled to 4 degrees C, a temperature that should induce the gel phase transition. Overall, these studies suggest that ANX12 is well suited to being a Ca2+ sensor for rapid all-or-none intercellular membrane-related events.     

Effect of cholesterol content on affinity and stability of factor VIII and annexin V binding to a liposomal bilayer membrane

To investigate the effect of cholesterol composition on the binding of factor VIII (FVIII) and annexin V (AV) to membranes, liposomal membranes with phospholipid bilayers of various compositions of phosphatidylcholine (PC), phosphatidylserine (PS), and cholesterol were constructed. A surface plasmon resonance (SPR) biosensor system was employed to measure the equilibrium and rate constants of the bindings. As expected, PS was found to play a dominant role in the binding of AV; its binding level was directly proportional to the PS composition in a liposome. The binding levels of FVIII and AV to liposome increased with an increase in cholesterol composition in liposome. It seemed to suggest that cholesterol in liposome acts as a ‘phospholipid arrangement’ factor by inducing the formation of PS-rich microdomains. However, in the absence of PS (20% on a mole basis), cholesterol could not exert the binding enhancement effect, which again confirmed the critical role of PS in the bindings. Stability of the AV binding was significantly improved by the increase in cholesterol content; for AV, the dissociation rate constant was decreased approximately fivefold, from 1.7 × 10^?3 s^?1 in the absence of cholesterol to 3.3 × 10^?4 s^?1 in the presence of only 10% cholesterol. But, for FVIII the binding stability was not so much influenced by the cholesterol addition (up to 50% on a mole basis). In summary, by using liposomes on an SPR system, we were able to demonstrate quantitatively the apparent effects of cholesterol on the binding affinity and stability of the membrane-binding proteins.     

Cholesterol modulates the membrane binding and intracellular distribution of annexin 6

Annexins are Ca(2+)- and phospholipid-binding proteins that are widely expressed in mammalian tissues and that bind to different cellular membranes. In recent years its role in membrane traffic has emerged as one of its predominant functions, but the regulation of its intracellular distribution still remains unclear. We demonstrated that annexin 6 translocates to the late endocytic compartment in low density lipoprotein-loaded CHO cells. This prompted us to investigate whether cholesterol, one of the major constituents of low density lipoprotein, could influence the membrane binding affinity and intracellular distribution of annexin 6. Treatment of crude membranes or early and late endosomal fractions with digitonin, a cholesterol-sequestering agent, displayed a strong reduction in the binding affinity of a novel EDTA-resistant and cholesterol-sensitive pool of annexin 6 proteins. In addition, U18666A-induced accumulation of cholesterol in the late endosomal compartment resulted in a significant increase of annexin 6 in these vesicles in vivo. This translocation/recruitment correlates with an increased membrane binding affinity of GST-annexin 6 to late endosomes of U18666A-treated cells in vitro. In conclusion, the present study shows that changes in the intracellular distribution and concentration of cholesterol in different subcellular compartments participate in the reorganization of intracellular pools of Ca(2+)-dependent and -independent annexin 6.     

Role of paired basic residues of protein C-termini in phospholipid binding

It is a well known phenomenon that the occurrence of several distinct amino acids at the C-terminus of proteins is non-random. We have analysed all Saccharomyces cerevisiae proteins predicted by computer databases and found lysine to be the most frequent residue both at the last (-1) and at the penultimate amino acid (-2) positions. To test the hypothesis that C-terminal basic residues efficiently bind to phospholipids we randomly expressed GST-fusion proteins from a yeast genomic library. Fifty-four different peptide fragments were found to bind phospholipids and 40% of them contained lysine/arginine residues at the (-1) or (-2) positions. One peptide showed high sequence similarity with the yeast protein Sip18p. Mutational analysis revealed that both C-terminal lysine residues of Sip18p are essential for phospholipid-binding in vitro. We assume that basic amino acid residues at the (-1) and (-2) positions in C-termini are suitable to attach the C-terminus of a given protein to membrane components such as phospholipids, thereby stabilizing the spatial structure of the protein or contributing to its subcellular localization. This mechanism could be an additional explanation for the C-terminal amino acid bias observed in proteins of several species.     

Competition of annexin V and anticardiolipin antibodies for binding to phosphatidylserine containing membranes

Annexin V, an intracellular protein with a calcium-dependent high affinity for anionic phospholipid membranes, acts as an inhibitor of lipid-dependent reactions of the blood coagulation. Antiphospholipid antibodies found in the plasma of patients with antiphospholipid syndrome generally do not interact with phospholipid membranes directly, but recognize (plasma) proteins associated with lipid membranes, mostly prothrombin or beta(2)-glycoprotein I (beta(2)GPI). Previously, it has been proposed that antiphospholipid antibodies may cause thrombosis by displacing annexin V from procoagulant cell surfaces. We used ellipsometry to study the binding of annexin V and of complexes of beta(2)GPI with patient-derived IgG antibodies to beta(2)GPI, commonly referred to as anticardiolipin antibodies (ACA), to phospholipid bilayers composed of phosphatidylcholine (PC) and 20% phosphatidylserine (PS). More specifically, we investigated the competition of these proteins for the binding sites at these bilayers. We show that ACA-beta(2)GPI complexes, adsorbed to PSPC bilayers, are displaced for more than 70% by annexin V and that annexin V binding is unaffected by the presence of ACA-beta(2)GPI complexes. Conversely, annexin V preadsorbed to these bilayers completely prevents adsorption of ACA-beta(2)GPI complexes, and none of the preadsorbed annexin V is displaced by ACA-beta(2)GPI complexes. Using ellipsometry, we also studied the effect of ACA-beta(2)GPI complexes on the interaction of annexin V with the membranes of ionophore-activated blood platelets as a more physiological relevant model of cell membranes. The experiments with blood platelets confirm the high-affinity binding of annexin V to these membranes and unequivocally show that annexin V binding is unaffected by the presence of ACA-beta(2)GPI. In conclusion, our data unambiguously show that ACA-beta(2)GPI complexes are unable to displace annexin V from procoagulant membranes to any significant extent, whereas annexin V does displace the majority of preadsorbed ACA-beta(2)GPI complexes from these membranes.     

Ceramide-platform formation and -induced biophysical changes in a fluid phospholipid membrane

To understand the formation and properties of ceramide-enriched domains in cell membranes, the biophysical properties of the binary system palmitoyloleoylphosphatidylcholine (POPC)/palmitoylceramide were thoroughly characterized. Diverse fluorescent probes and parameters were necessary to unravel the complexity of this apparently simple system. For the first time, a complete phase diagram is reported, characterizing the lamellar phases of these mixtures, and providing a quantitative framework integrating biophysical and biological studies. The diagram suggests that in resting cells no ceramide domains exist, but upon apoptotic stimuli, platforms may form. Moreover, our data show that 2 mol% of Cer strongly affects the POPC fluid matrix, suggesting that a small increase in Cer levels can significantly affect cell membrane properties. In this work, we also show that Cer domains, formed in conditions similar to physiological, are extremely ordered and rigid. The domains composition is estimated from the phase diagram and their large size was concluded from fluorescence resonance energy transfer. Dynamic light scattering and electron microscopy were used to characterize the system morphology, which is highly dependent on ceramide content and includes vesiculation and tubular structure formation.     

The cytoplasmic domains of phospholamban and phospholemman associate with phospholipid membrane surfaces

Phospholamban (PLB) and phospholemman (PLM, also called FXYD1) are small transmembrane proteins that interact with P-type ATPases and regulate ion transport in cardiac cells and other tissues. This work has investigated the hypothesis that the cytoplasmic domains of PLB and PLM, when not interacting with their regulatory targets, are stabilized through associations with the surface of the phospholipid membrane. Peptides representing the 35 C-terminal cytoplasmic residues of PLM (PLM(37-72)), the 23 N-terminal cytoplasmic residues of PLB (PLB(1-23)), and the same sequence phosphorylated at Ser-16 (P-PLB(1-23)) were synthesized to examine their interactions with model membranes composed of zwitterionic phosphatidylcholine (PC) lipids alone or in admixture with anionic phosphatidylglycerol (PG) lipids. Wide-line 2H NMR spectra of PC/PG membranes, with PC deuterated in the choline moiety, indicated that all three peptides interacted with the membrane surface and perturbed the orientation of the choline headgroups. Fluorescence and 31P magic-angle spinning (MAS) NMR measurements indicated that PLB(1-23) and P-PLB(1-23) had a higher affinity for PC/PG membranes, which carry an overall negative surface charge, than for PC membranes, which have no net surface charge. The 31P MAS NMR spectra of the PC/PG membranes in the presence of PLM(37-72), PLB(1-23), and P-PLB(1-23) indicated that all three peptides induced clustering of the lipids into PC-enriched and PG-enriched regions. These findings support the theory that the cytoplasmic domains of PLB and PLM are stabilized by interacting with lipid headgroups at the membrane surface, and it is speculated that such interactions may modulate the functional properties of biological membranes.     

Membrane-bound annexin V isoforms (CaBP33 and CaBP37) and annexin VI in bovine tissues behave like integral membrane proteins.

The distribution of annexin V isoforms (CaBP33 and CaBP37) and of annexin VI in bovine lung, heart, and brain subfractions was investigated with special reference to the fractions of these proteins which are membrane-bound. In addition to EGTA-extractable pools of the above proteins, membranes from lung, heart, and brain contain EGTA-resistant annexins V and VI which can be solubilized with detergents (Triton X-100 or Triton X-114). A strong base like Na2CO3, which is usually effective in extracting membrane proteins, only partially solubilizes the membrane-bound, EGTA-resistant annexins analyzed here. Also, only 50-60% of the Triton X-114-soluble annexins partition in the aqueous phase, the remaining fractions being recovered in the detergent-rich phase. Altogether, these findings suggest that, by an as yet unknown mechanism, following Ca(2+)-dependent association of annexin V isoforms and annexin VI with membranes, substantial fractions of these proteins remain bound to membranes in a Ca(2+)-independent way and behave like integral membrane proteins. These results further support the possibility that the above annexins might play a role in membrane trafficking and/or in the regulation of the structural organization of membranes.     

Expression and purification of recombinant annexin V for the detection of membrane alterations on apoptotic cells

Apoptosis is a mode of cell death that is accompanied by specific alterations to the plasma membrane that promote the recognition and engulfment of these cells by phagocytes. Although several such membrane alterations have been defined, redistribution of phosphatidylserine from the inner to the outer plasma membrane leaflet has become one of the most widely used markers for apoptotic cells in mammals. This is largely due to the availability of a sensitive and specific probe for this event in the form of the phosphatidylserine-binding protein, annexin V. Here, we describe methods for the expression and purification of recombinant polyhistidine-tagged annexin V from Escherichia coli. Recombinant annexin V is highly soluble and is thus readily expressed and purified to high yields; typically in the region of 4microg of protein per ml of bacterial culture. We also describe methods for conjugation of this protein to the FITC fluorophore and for its use for the detection of apoptotic cells by flow cytometry or fluorescence microscopy.     

Influence of membrane components in the binding of proteins to membrane surfaces

We have quantified the enhancement of membrane binding of activated and deactivated Galpha(s) and Galpha(q) subunits, Gbetagamma subunits, and phospholipase Cbeta(2) by lipid rafts and by the presence of membrane-associated protein partners. Membrane binding studies show that lipid rafts do not affect the intrinsic membrane affinity of Galpha(q)(GDP) and Galpha(s)(GDP), supporting the idea that these proteins partition evenly between the domains. Visualization of lipid rafts on monolayers by use of a probe that does not enter raft domains shows that neither activated nor deactivated Galpha(q)(GDP) subunits distribute evenly between the raft and nonraft domains, contrary to previous suggestions. Membrane binding of deactivated Galpha(q) and Galpha(s)(GDP) became weaker when Gbetagamma subunits were present, in contrast with the behavior predicted by thermodynamics. However, activated Galpha subunits and phospholipase Cbeta(2) were recruited to membrane surfaces by protein partners by predicted amounts. Our studies suggest that the anomalous behavior seen for deactivated Galpha subunits in the presence of Gbetagamma subunits may be due to conformational changes in the N-terminus and/or occlusion of a portion of its membrane interaction region by Gbetagamma. Even though membrane recruitment was clearly observed for one protein partner, the presence of a second partner of lower affinity did not further promote membrane binding. For these proteins, the formation of larger protein complexes with very high membrane affinities is unlikely.     

Conformational change of mastoparan from wasp venom on binding with phospholipid membrane

Aglycosylated IgG produced by hybridoma cells cultured in the presence of tunicamycin was compared with normal IgG for its ability to bind to staphylococcal protein A. No differences were found in binding or elution profiles. It is concluded that aglycosylation does not produce major structural alterations at the CH2-CH3 interface of the Fc region of IgG.     

Molecular dynamics of a vasopressin V2 receptor in a phospholipid bilayer membrane

Molecular dynamics simulations were carried out for a V2 receptor (V2R) model embedded in a dimyristoylphosphatidylcholine (DMPC) bilayer. Both free and ligand-bound states of V2R were modeled. Our initial V2R model was obtained using a rule-based automated method for GPCR modeling and refined using constrained simulated annealing in vacuo. The docking site of the native vasopressin ligand was selected and justified upon consideration of ligand-receptor interactions and structure-activity data. The primary purpose of this work was to investigate the usefulness of MD simulation of an integral membrane protein like a GPCR receptor, upon inclusion of a carefully parameterized surrounding lipid membrane and water. Physical properties of the system were evaluated and compared with the fully hydrated pure DMPC bilayer membrane. The solvation interactions, individual lipid-protein interaction and fluctuations of the protein, the lipid, and water were analyzed in detail. As expected, the membrane-spanning helices of the protein fluctuate less than the peripheral loops do. The protein appears to disturb the local lipid structure. Simulations were carried out using AMBER 4.1 package upon constant number-pressure-temperature (NPT) conditions on massively parallel computers Cray T3E and IBM SP2.     

Promotion of acid-induced membrane fusion by basic peptides. Amino acid and phospholipid specificities

The ability of oligo- and polymers of the basic amino acids L-lysine, L-arginine, L-histidine and L-ornithine to induce lipid intermixing and membrane fusion among vesicles containing various anionic phospholipids has been investigated. Among vesicle consisting of either phosphatidylinositol or mixtures of phosphatidic acid and phosphatidylethanolamine rapid and extensive lipid intermixing, but not complete fusion, was induced at neutral pH by poly-L-ornithine or L-lysine peptides of five or more residues. When phosphatidylcholine was included in the vesicles, the lipid intermixing was severely inhibited. Such lipid intermixing was also much less pronounced among phosphatidylserine vesicles. Poly-L-arginine provoked considerable leakage from the various anionic vesicles and caused significantly less lipid intermixing than L-lysine peptides at neutral pH. When the addition of basic amino acid polymer was followed by acidification to pH 5-6, vesicle fusion was induced. Fusion was more pronounced among vesicles containing phosphatidylserine or phosphatidic acid than among those containing phosphatidylinositol, and occurred also with vesicles whose composition resembles that of cellular membranes (i.e., phosphatidylcholine/phosphatidylethanolamine/phosphatidylserine, 50:30:20, by mol). Liposomes with this composition are resistant to fusion by Ca2+ or by acidification after lectin-mediated contact. The tight interaction among vesicles at neutral pH, resulting in lipid intermixing, does not seem to be necessary for the fusion occurring after acidification, but the basic peptides nevertheless appear to play a more active role in the fusion process than simply bringing the vesicles in contact. However, protonation of the polymer side chains and transformation of the polymer into a polycation does not explain the need for acidification, since the pH-dependence was quite similar for poly(L-histidine)- and poly(L-lysine)-mediated fusion.