A microassay for the pore-forming activity of complement, perforin, and other cytolytic proteins based on confocal laser scanning microscopy

A fluorescence microscopic assay for the activity of complement, perforin, and other cytolytic proteins which form transmembrane pores in cellular membranes is described. The assay was worked out and tested with red blood cell membranes (ghosts) and was then applied to intact hemoglobin-free cells. Resealed human erythrocyte ghosts were incubated with complement or perforin. A small polar fluorescent probe (fluorescein-labeled 1-kDa dextran, FD1) which permeates through complement and perforin pores but not through normal cell membranes was added to the samples. The capability of the confocal laser scanning microscope (CLSM) to generate thin optical sections was exploited to visualize and quantitate fluorescence inside single ghosts and thus determine the fraction of ghosts which had become permeable for FD1. The activity of complement or perforin was quantitated by plotting the fraction of permeable cells versus the concentration of the pore-forming protein. The results were in good agreement with those of a conventional hemolytic assay. The CLSM-based assay was then applied to intact hemoglobin-free cells for which only few alternative assays are available. Compared to conventional hemolytic assays for the activity of pore-forming proteins the assay described here can be applied to a large variety of natural and artificial membrane systems. The assay can be performed under nonlysing conditions. Furthermore, the assay is simple, relatively fast, and requires only extremely small amounts of cells and pore-forming proteins.     

Human erythrocyte galactosyltransferase. Characterization, membrane association and sidedness of active site

Human erythrocyte UDPgalactose : 2-acetamido-2-deoxy-α-d-galactopyranosylpeptide galactose transferase (Galactosyltransferase) has been characterized in terms of detergent and metal ion requirements, Michaelis constants for donor and acceptor substrates, inhibition constant for N-acetylgalactosamine, pH optimum and ionic strength effects. The assay thus optimized permits initial velocity measurements. Galactosyltransferase was shown to be membrane-bound by demonstrating its association with erythrocyte ghosts after high and low ionic strength treatments to remove weakly-associated proteins. In the absence of detergents, no activity was detectable in sealed ghosts and inside-out vesicles derived from erythrocyte membranes. Enzyme activation by detergents paralleled solubilization of membrane proteins. Both latency and solubilization studies indicated a substrate-inaccessible active site for the enzyme in situ in the membrane. Galactosyltransferase activity in resealed ghosts, leaky ghosts and inside-out vesicles was resistant to the action of trypsin, chymotrypsin or pronase applied as single agents. A mixture of these proteases, however, strongly reduced the enzyme activity in inside-out vesicles and leaky ghosts, indicating a cytosolic orientation for the active site of the galactosyltransferase.     

The sub-membrane reticulum of the human erythrocyte: A scanning electron microscope study

A web-like reticulum underlying the human erythrocyte membrane was studied at a resolution of 5–10 nm by means of a scanning electron microscope. The network was visualized in isolated membranes (ghosts) torn open to reveal their interior space and in residues derived from ghosts extracted with Triton X-100. It formed a continuous (rather than patchy) cover over the entire cytoplasmic surface, except where lifted off or torn away. Filaments (5–40 nm in diameter), annular figures (40–60 nm in diameter), and nodes (30–100 nm in diameter) were prominent in different networks. The dimensions of the filaments and the interstices in the reticulum varied with conditions, suggesting that the network has elastic properties. This reticulum is probably related to the erythrocyte membrane proteins spectrin and actin.     

Solubilization, reconstitution, and attempted affinity chromatography of the sugar transporter of the erythrocyte membrane

Reconstitution of the sugar transport system of human erythrocytes into artificial liposomes was achieved by freezing, thawing, and sonicating preformed phospholipid vesicles in the presence of intact ghosts, protein-depleted ghosts, or detergent-treated ghosts. D-glucose equilibrium exchange activities and affinity constants in the range of the reported erythrocyte values were reached in the best experiments. Whereas the extraction of peripheral membrane proteins did not depress the transport function crucially after reconstituting these protein-depleted ghosts, the selective solubilization of integral membrane proteins by a variety of nonionic detergents resulted in an uncontrollable, continuously increasing inactivation of the carrier. However, Emulphogene BC-720 extracts could be prepared in which the glucose transporter retained activity for days at 4 degrees C. These extracts were applied to affinity chromatography matrices of phloretin-Agarose, prepared by coupling phloretinyl-3'-benzylamine (PBA) to CH-Sepharose 4B and to Affigel 202. Although the solubilized sugar transporter appeared to be selectively adsorbed to both PBA matrices, it could not be eluted by specific counter ligands or gentle eluants in a biologically active form. However, chaotropic agents could be used to elute intrinsic proteins, including bands 3 and 4.5, from the Affigel affinity medium.     

Effect of complement on the lateral mobility of erythrocyte membrane proteins. Evidence for terminal complex interaction with cytoskeletal components

The lateral mobilities of erythrocyte membrane proteins and terminal complement complexes (TCC) were measured on C-treated erythrocyte ghosts by the technique of fluorescence redistribution after photobleaching. Results showed that the lateral diffusion coefficient of the bulk membrane proteins decreased with the assembly of TCC on the membrane at low C dose and was significantly reduced with assembly of the full membrane attack complex (C5b-9), even in the absence of cell lysis. At high serum doses, the mobility of the membrane proteins increased slightly above that of the control cells. The diffusion coefficients of the TCC on the erythrocyte membrane range from 1.18 to 4.37 x 10(-11) cm2/s, values characteristic of anchored membrane proteins. Spectrin-depletion of the C-lysed erythrocytes results in 25- and 45-fold increases in the diffusion coefficients of the membrane proteins and the C5b-9 complex, respectively. Conversely, oxidative cross-linking of spectrin by diamide reduced the diffusion coefficients of both membrane and C proteins. These studies indicate that the deposition of TCC on an erythrocyte can result in a substantial change in the physical and structural properties of the target membrane, aside from the creation of functional lesions. The low mobilities of the terminal complexes on the target membrane suggest possible interactions with cytoskeletal elements or with anchored membrane proteins.     

Cell volume and osmotic properties of erythrocytes after complement lysis measured by flow cytometry

The changes of volume distribution curves of erythrocytes during and after lysis by complement or nystatin or in hypotonic buffers were measured by flow cytometry. Biconcave and spheroidal ghosts were observed after complement lysis and spheroidal ghosts were seen only after nystatin and hypotonic lysis. The spheroidal ghosts derived from red cells lysed by complement or nystatin were permeable to sucrose; those from hypotonic lysis were sucrose-impermeable. Spheroidal ghosts after complement lysis remained permeable for sucrose whereas spheroidal ghosts after nystatin lysis resealed after removal of the drug by washing. Biconcave ghosts produced by complement lysis were almost impermeable to sucrose initially and therefore responded to osmotic changes, but they became sucrose-permeable upon prolonged incubation at 37 degrees C. The rate of sucrose equilibration increased as the stability of the biconcave shape diminished with increasing numbers of C5b-9 complexes. At 850 C5b-9 complexes/ghost, the biconcave shape and impermeability for sucrose were completely lost. The results support the hypothesis that complement C5b-9 complexes, in addition to the interaction with the lipid bilayer, may interact with the cytoskeleton of the erythrocyte membrane.     

Sendai Virus Fusion Activity as Modulated by Target Membrane Components

We have studied the differences between erythrocytes and erythrocyte ghosts as target membranes for the study of Sendai virus fusion activity. Fusion was monitored continuously by fluorescence dequenching of R18-labeled virus. Experiments were carried out either with or without virus/target membrane prebinding. When Sendai virus was added directly to a erythrocyte/erythrocyte ghost suspension, fusion was always lower than that obtained when experiments were carried out with virus already bound to the erythrocyte/erythrocyte ghost in the cold, since with virus prebinding fusion can be triggered more rapidly. Although virus binding to both erythrocytes and erythrocyte ghosts was similar, fusion activity was much more pronounced when erythrocyte ghosts were used as target membranes. These observations indicate that intact erythrocytes and erythrocyte ghosts are not equivalent as target membranes for the study of Sendai virus fusion activity. Fusion of Sendai virus with both target membranes was inhibited when erythrocytes or erythrocyte ghosts were pretreated with proteinase K, suggesting a role of target membrane proteins in this process. Treatment of both target membranes with neuraminidase, which removes sialic acid residues (the biological receptors for Sendai virus) greatly reduced viral binding. Interestingly, this treatment had no significant effect on the fusion reaction itself.     

The influence of electrochemical gradients of Na+ and K+ upon the membrane binding and pore forming activity of the terminal complement proteins

Summary The hemolytic activity of the terminal complement proteins (C5b-9) towards erythrocytes containing high potassium concentration has been reported to be dramatically increased when extracellular Na⁺ is substituted isotonically by K⁺ (Dalmasso, A.P., et al., 1975,J. Immunol. 115:63–68). This phenomenon was now further investigated using resealed human erythrocyte ghosts (ghosts), which can be maintained at a nonlytic osmotic steady state subsequent to C5b-9 binding: (1) The functional state of C5b-9-treated ghosts was studied from their ability to retain trapped [¹⁴C]-sucrose or [³H]-inulin when suspended either in the presence of Na⁺ or K⁺. A dramatic increase in the permeability of the ghost membrane to both nonelectrolytes-in the absence of significant hemoglobin release-was observed for C5b-9 assembly in the presence of external K⁺. (2) The physical binding of the individual¹²⁵I-labeled terminal complement proteins to ghost membranes was directly measured as a function of intra- and extracellular K⁺ and Na⁺. The uptake of¹²⁵I-C7,¹²⁵I-C8, and¹²⁵I-C9 into membrane C5b-9 was unaltered by substitution of Na⁺ by K⁺. (3) The binding of the terminal complement proteins to ghosts subjected to a transient membrane potential generated by the K⁺-ionophore valinomycin (in the presence of K⁺ concentration gradients) was measured. No significant change in membrane binding of any of the C5b-9 proteins was detected under the influence of both depolarizing and hyperpolarizing membrane potentials. It can be concluded that the differential effect of Na⁺ versus K⁺ upon the erythrocyte membrane isnot due to an effect upon the binding of the complement proteins to the membraneper se, but upon the functional properties of the assembled C5b-9 pore site.     

Spatial relationships of the proteins of vesicular stomatitis virus: induction of reversible oligomers by cleavable protein cross-linkers and oxidation.

To delineate the proximity and spatial arrangement of the major structural proteins of intact vesicular stomatitis (VS) virions, protein complexes formed by oxidation or by bivalent cross-linkers were analyzed by two-dimensional electrophoresis on polyacrylamide slab gels. H2O2 oxidation of VS virions produced an N-polypeptide dimer (molecular weight, approximately equal to 110,000) on a first dimension gel that could be reduced to N monomers (molecular weight, approximately equal to 50,000). Proteins extracted from unreduced and unoxidized VS virions contained dimeric and trimeric forms of M-protein complexes as well as a heterodimer of M and N protein. Qualitatively similar VS viral protein complexes were generated by exposing VS virions to the reversible protein cross-linkers methyl-4-mercaptobutyrimidate (MMB), tartryl diazide (TDA), and dithiobis(succinimidyl proprionate) (DTBSP); cross-linked complexes on first-dimension gels were cleaved by reduction with 2-mercaptoethanol (MMB or DTBSP cross-linked) or by periodate oxidation (TDA cross-linked). In addition to covalently linked homodiamers of M and N proteins and a protein M-N heterodimer, the protein cross-linkers also generated homo-oligomers of G protein and a G-M heterodimer. These data suggest that the glycoprotein spike of VS virus is composed of more than one G protein. The existence of N-M and G-M heterodimers is consistent with the hypothesis that the matrix (M) protein may serve as a bridge between the G and N proteins in assembly of the VS virion.     

Involvement of cytoskeletal proteins in the barrier function of the human erythrocyte membrane. III. Permeability of spectrin-depleted inside-out membrane vesicles to hydrophilic nonelectrolytes. Formation of leaks by chemical or enzymatic modification of membrane proteins.

Spectrin-depleted inside-out vesicles (IOV's) prepared from human erythrocyte membranes were characterized in terms of size, ground permeability to hydrophilic nonelectrolytes and their sensitivity to modification by SH reagents, DIDS and trypsin. IOV's proved to have the same permeability of their lipid domain to erythritol as native erythrocytes, in contrast to resealed ghosts (Klonk, S. and Deuticke, B. (1992) Biochim. Biophys. Acta 1106, 126-136 (Part I in this series)), which have a residual leak. On the other hand, IOV's have a slightly elevated permeability for mannitol and sucrose, nonelectrolytes which are almost (mannitol) or fully (sucrose) impermeant in the native membrane. These increased fluxes, which have a high activation energy and can be stimulated by phloretin, are, however, also much smaller than the corresponding leak fluxes observed in resealed ghosts. In view of these differences, formation of IOV's can be concluded to go along with partial annealing of barrier defects persisting in the erythrocyte membrane after preparation of resealed ghosts. Oxidation of SH groups of the IOV membrane by diamide produces an enhancement of permeability for hydrophilic nonelectrolytes which is much less pronounced than that induced by a similar treatment of erythrocytes or ghosts (Klonk, S. and Deuticke, B. (1992) Biochim. Biophys. Acta 1106, 126-136 (Part I in this series)). Moreover, proteolytic treatment of the vesicle membrane, although leading to a marked digestion of integral membrane proteins, only induces a minor, saturating increase of permeability, much lower than that in trypsinized resealed ghosts (Klonk, S. and Deuticke, B. (1992) Biochim. Biophys. Acta 1106, 137-142 (Part II of this series)). Since absence of the cytoskeletal proteins, spectrin and actin, is the major difference between IOV's and resealed ghosts, these results may be taken as further evidence for a dependence of the barrier properties of the erythrocyte membrane bilayer domain on its interaction with cytoskeletal elements. In contrast, these barrier properties seem to be rather insensitive to perturbations of integral proteins.     

Photodynamic action of bilirubin on human erythrocyte membranes. Modification of polypeptide constituents.

The photodynamic action of bilirubin on isolated human erythrocyte membranes (ghosts) has been studied. When incorporated into ghosts (pH 8.0,10 degrees) the bile pigment photosensitizes in blue light the peroxidation of unsaturated lipids, as evidenced by a positive color reaction with 2-thiobarbituric acid. Accompanying lipid peroxidation was the disappearance of most of the major membrane proteins (Coomassie Blue staining in sodium dodecyl sulfate-polyacrylamide gel electrophoresis) and appearance of polypeptide photoproducts of greater size (mol wt greater than 250,000). The association of membrane proteins (presumably by cross-linking) was insignificant when bilirubin-ghost suspensions were kept in the dark, or when ghosts were irradiated in the absence of bilirubin. Electrophoretic bands 1 and 2 (Fairbanks, G., Steck, T.L., and Wallach, D. F.H (1971), Biochemistry 10, 2606) diminished rapidly during the photoreaction, whereas band 3 and the three sialoglycoproteins disappeared at a much slower rate. Dispersal of membrane consituents by treatment with sodium dodecyl sulfate prior to irradiation resulted in relatively little peroxidation and no noticeable formation of high molecular weight polypeptide complexes. The possibility that malonaldehyde, a product of lipid peroxidation, is involved in cross-linking during irradiation was studied by incubating ghosts with exogenous malonaldehyde. Although the reagent did cross-link membrane proteins (electrophoretic bands 1, 2, 2.1 2.2, and 4.1 diminished most rapidly and high molecular weight bands appeared), the reaction could only be demonstrated with malonaldehyde concentrations several orders of magnitude greater than those detected in irradiation experiments. If malonaldehyde cross-linking occurs, it does not appeare to be the predominant mechanism of polypeptide association during irradiation of bilirubin-containing ghosts.     

Human erythrocyte galactosyltransferase. Characterization, membrane association and sidedness of active site

Human erythrocyte UDPgalactose : 2-acetamido-2-deoxy-alpha-D-galactopyranosylpeptide galactose beta(1 lead to 3) transferase (Galactosyltransferase) has been characterized in terms of detergent and metal ion requirements. Michaelis constants for donor and acceptor substrates, inhibition constant for N-acetylgalactosamine, pH optimum and ionic strength effects. The assay thus optimized permits initial velocity measurements. Galactosyltransferase was shown to be membrane-bound by demonstrating its association with erythrocyte ghosts after high and low ionic strength treatments to remove weakly-associated proteins. In the absence of detergents, no activity was detectable in sealed ghosts and inside-out vesicles derived from erythrocyte membranes. Enzyme activation by detergents paralleled solubilization of membrane proteins. Both latency and solubilization studies indicated a substrate inaccessible active site for the enzyme in situ in the membrane. Galactosyltransferase activity in resealed ghosts, leaky ghosts and inside-out vesicles was resistant to the action of trypsin, chymotrypsin or pronase applied as single agents. A mixture of these proteases, however, strongly reduced the enzyme activity in inside-out vesicles and leaky ghosts, indicating a cytosolic orientation for the active site of the galactosyltransferase.     

Structure and thermotropic phase behaviour of detergent-resistant membrane raft fractions isolated from human and ruminant erythrocytes

Detergent-resistant membrane raft fractions have been prepared from human, goat, and sheep erythrocyte ghosts using Triton X-100. The structure and thermotropic phase behaviour of the fractions have been examined by freeze-fracture electron microscopy and synchrotron X-ray diffraction methods. The raft fractions are found to consist of vesicles and multilamellar structures indicating considerable rearrangement of the original ghost membrane. Few membrane-associated particles typical of freeze-fracture replicas of intact erythrocyte membranes are observed in the fracture planes. Synchrotron X-ray diffraction studies during heating and cooling scans showed that multilamellar structures formed by stacks of raft membranes from all three species have d-spacings of about 6.5 nm. These structures can be distinguished from peaks corresponding to d-spacings of about 5.5 nm, which were assigned to scattering from single bilayer vesicles on the basis of the temperature dependence of their d-spacings compared with the multilamellar arrangements. The spacings obtained from multilamellar stacks and vesicular suspensions of raft membranes were, on average, more than 0.5 nm greater than corresponding arrangements of erythrocyte ghost membranes from which they were derived. The trypsinization of human erythrocyte ghosts results in a small decrease in lamellar d-spacing, but rafts prepared from trypsinized ghosts exhibit an additional lamellar repeat 0.4 nm less than a lamellar repeat coinciding with rafts prepared from untreated ghosts. The trypsinization of sheep erythrocyte ghosts results in the phase separation of two lamellar repeat structures (d = 6.00; 5.77 nm), but rafts from trypsinized ghosts produce a diffraction band almost identical to rafts from untreated ghosts. An examination of the structure and thermotropic phase behaviour of the dispersions of total polar lipid extracts of sheep detergent-resistant membrane preparations showed that a reversible phase separation of an inverted hexagonal structure from coexisting lamellar phase takes place upon heating above about 30 °C. Non-lamellar phases are not observed in erythrocytes or detergent-resistant membrane preparations heated up to 55 °C, suggesting that the lamellar arrangement is imposed on these membrane lipids by interaction with non-lipid components of rafts and/or that the topology of lipids in the erythrocyte membrane survives detergent treatment.     

Complement pores in erythrocyte membranes: Analysis of C8/C9 binding required for functional membrane damage

The number of membrane-bound terminal complement proteins (C5b-9) required to generate a functional pore in the human erythrocyte membrane ghost has been determined. Resealed erythrocyte ghost membranes (ghosts) were treated with human complement proteins C5b6, C7, ¹³¹I-C8, and ¹²⁵I-C9 under non-lytic conditions. Following C5b-9 assembly, sucrose-permeant ghosts were separated from C5b-9 ghosts that remained impermeant to sucrose by centrifugation over density barriers formed of 43% (w/v) sucrose. Analysis of ¹³¹I-C8 and ¹²⁵I-C9 bound to sucrose-permeant and sucrose-impermeant subpopulations of C5b-9 ghosts revealed: 1. Sucrose-permeant C5b-9 ghosts show increased uptake of both ¹³¹I-C8 and ¹²⁵I-C9 as compared to ghosts that remain impermeant to sucrose. Ghosts with less than 300 molecules ¹³¹I-C8 bound remain impermeant to sucrose, irrespective of the total C9 input, or, the multiplicity of C9 uptake by membrane C5b-8. 2. In the presence of excess ¹²⁵I-C9, the ratio of ¹²⁵I-C9/¹³¹I-C8 bound to membrane C5b67 is 3.2 ± 0.8 (mean ± 2 S.D.), suggesting an average stoichiometry of 3 C9 per C5b-8. Under these conditions, the ratio of ¹²⁵I-C9/¹³¹I-C8 bound to sucrose-permeant ghosts (3.3 ± 0.7) does not significantly differ from the ratio bound to sucrose-impermeant ghosts (2.9 ± 0.6). 3. With limiting C9 input, the threshold of total C5b-8 uptake required for sucrose permeability increases significantly above 300 per cell when the ratio of bound ¹²⁵I-C9/¹³¹I-C8 is decreased below unity. In the complete absence of C9, 11 700 C5b-8 complexes are bound to sucrose-permeant ghosts. It is concluded that more than 300 C5b-9 complexes must bind to the human erythrocyte to form a sucrose-permeant lesion. Although the binding of one C9 per C5b-8 is critical to the pore-forming activity of these proteins, the binding of additional molecules of C9 to each complex (C9/C8 > 1) does not significantly alter the threshold of total C5b-9 uptake required for lesion formation.     

Spin-label studies of the oligomeric structure of band 3 protein in erythrocyte membranes and in reconstituted systems

A spin-labeled fatty acid (16-doxylstearic acid), linked by an ester bond to a maleimide or a nitrene residue, was covalently attached to band 3 of erythrocyte membranes. The electron spin resonance spectrum of the spin-labeled protein was examined at different temperatures in: (a) whole erythrocyte ghosts; (b) ghosts depleted of spectrin and actin; (c) alkaline-treated ghosts; (d) vesicles made with purified band 3 reassociated with dimyristoylphosphatidylcholine. Most spectra are composite with a major component corresponding to a large overall splitting. The determination of the percentage of the immobilized component was carried out by pairwise subtraction. At low temperatures (1–7°C), the highest fraction of immobilized component was found in dimyristoylphosphatidylcholine vesicles (approx. 100%); alkaline-treated membranes had approx. 75% of the immobilized component at the same temperature; whole erythrocyte, spectrin/actin-depleted and spectrin/actin/ankyrin-depleted ghosts gave identical results (approx. 60% of immobilized component). The immobilized fraction decreased in all samples with increasing temperature or addition of a nonsolubilizing concentration of dodecyl octaethylene glycol monoether. In dimyristoylphosphatidylcholine vesicles, however, the modification in the ratio of the two components was obtained only above the lipid transition temperature (23°C). The strong immobilization of the spin-labeled lipid chain at all temperatures suggested trapping of the lipid chain between proteins. At low temperature, in dimyristoylphosphatidylcholine vesicles or in alkaline-treated ghosts, lipid-protein segregation is likely to take place. In whole erythrocyte ghosts, on the other hand, the large contribution of the motionally restricted component at physiological temperature indicates the oligomeric nature of band 3. Partial dissociation of the oligomers occurs as the temperature is increased, but the presence or absence of cytoskeletal proteins has no influence on the state of oligomerization of band 3.     

Binding of vanadate to human erythrocyte ghosts and subsequent events

Spectroscopic techniques were used to investigate the interaction between vanadate and human erythrocyte ghosts. Direct evidence from ⁵¹V nuclear magnetic resonance (NMR) studies suggested that the monomeric and polymeric vanadate species may bind to the anion binding sites of band 3 protein of the erythrocyte membrane. The results of ⁵¹V NMR studies and the quenching effect of vanadate on the intrinsic fluorescence of the membrane proteins indicated that in the low concentration range of vanadate (<0.6 mm), monomeric vanadate binds mostly to the anion sites of band 3 protein with the dissociation constant close to 0.23 mm. The experiments of sulfhydryl content titration by the method of Ellman and residue sulfhydryl-labeled fluorescence spectroscopies clearly displayed that vanadate reacts directly with sulfhydryl groups. The appearance of the anisotropic election spin resonance (ESR) signal of vanadyl suggests that a small (c. 3%) amount of vanadate was reduced by sulfhydryl groups of membrane proteins. The fluidity and order of intact ghost membrane were reduced by the reaction with vanadate, as shown by the ESR studies employing the protein- and lipid-specific spin labels. It was concluded that although vanadates mainly bind to band 3 protein, a minor part of vanadate may oxidize the residue sulfhydryl groups of membrane proteins, and thus decrease the fluidity of erythrocyte membrane.     

Rotational diffusion of band 3 in erythrocyte membranes. 1. Comparison of ghosts and intact cells.

The rotational diffusion of eosin-labeled 3 in human erythrocyte cells and hemoglobin-free ghosts at 37 degrees C has been studied in detail by polarized delayed luminescence. The time-resolved anisotropy with both cells and freshly prepared ghosts is similar, decaying with well-resolved rotational correlation times of 0.03, 0.2, and greater than or equal to 1 ms. Mild proteolytic removal of the water-soluble 41-kDa cytoplasmic domain of band 3 in ghosts results in a drastic increase in the fractional contributions of the two fastest depolarizing components. Our results, taken together with other data in the literature, imply that several classes of band 3 that differ greatly in mobility exist in ghosts and intact cells. The mobility of one class is hindered due to complexation with other membrane or cytoplasmic proteins mediated via the 41-kDa cytoplasmic domain. However, another class of band 3 molecules exists as homo-or heterooligomeric complexes larger than a dimer that are stabilized by hydrophobic interactions involving the intramembranal domain. Finally, the presence of the (previously undetected) 0.03-ms anisotropy component strongly suggests that a significant fraction of band 3 in both ghosts and intact cells is highly mobile and diffuses at the rate expected for a freely rotating dimer in the erythrocyte membrane.     

The formation of vesicles retaining sodium-dependent transport systems for amino acids from protein-depleted membranes of pigeon erythrocytes

The process of the formation of vesicles from pigeon erythrocyte membranes was studied. Mildly alkaline solutions of low ionic strength, which reduce human erythrocyte membranes to small vesicles depleted of spectrin and other proteins, have no such effect on pigeon erythrocyte ghosts. A distinct phase of removal of membrane proteins, including spectrin, began to occur only when pigeon erythrocyte membranes were exposed to 0.2 mM EDTA adjusted to pH values above 10.2. Vesicles which demonstrated Na⁺-dependent amino acid transport were generated between the pH values 10.8 and 11.4. The results show that peripheral proteins, notably spectrin, maintain the integrity of the pigeon erythrocyte ghost. The interaction of these proteins with the membrane is rather different from that well studied in the human erythrocyte ghost and the possible significance of this for the pigeon erythrocyte is discussed.     

Strong interactions between a spin-labeled cholesterol analog and erythrocyte proteins in the human erythrocyte membrane

We have used a spin label analog of cholesterol bearing a nitroxide on the alkyl chain (26-nor-25-doxylcholestanol) to study cholesterol-protein interactions in the human erythrocyte membrane. As judged from the ESR spectrum, the spin label is readily incorporated into the membrane when added from a concentrated ethanolic solution to a cell or ghost suspension. With intact erythrocytes or white ghosts in isotonic buffer, the ESR spectrum is a superposition of a mobile component and a strongly immobilized component (outer hyperfine splitting 61–63 G). The latter corresponds to approx. 45% of the signal, a percentage which is barely affected by varying the temperature between 5 and 37°C. Removal of the cytoskeletal proteins spectrin and actin by low ionic strength treatment or of all extrinsic proteins by alkali treatment of ghosts reduces the immobilized fraction to approx. 25%. The effect of controlled proteolysis of intrinsic proteins was also tested. Pre-treatment of cells with chymotrypsin or pre-treatment of unsealed ghosts with trypsin has no effect on the ESR spectrum obtained with alkali-treated membranes. On the other hand, after chymotrypsin treatment of unsealed ghost, which reduces the band 3 protein to a 17.5 kDa membrane fragment, the strongly immobilized component is no longer observable. These data show that the cholesterol analog 26-nor-25-doxylcholestanol interacts strongly with one or several proteins of the erythrocyte membrane. That the intrinsic protein band 3 is involved is suggested by the disappearance of the immobilized fraction occurring upon chymotrypsin digestion of this protein. Our results are thus consistent with the proposal of a selective cholesterol-band 3 interaction in the erythrocyte membrane (Schubert, D. and Boss, K. (1982) FEBS Lett. 150, 4–8). Our data also suggest that this interaction is influenced by cytoskeletal proteins, an effect which can be explained considering the known linking of band 3 to the erythrocyte cytoskeleton via ankyrin. Experiments have also been carried out with 3-doxylandrostanol, a more commonly used cholesterol spin-label analog. With this spin label, at all temperatures investigated, we found it impossible to demonstrate unambiguously the existence of two separate spectral components. It is suggested that 26-nor-25-doxylcholestanol is a better reporter of cholesterol behavior in membranes.     

Limited proteolysis of the erythrocyte membrane skeleton by calcium-dependent proteinases

The action of purified calcium-dependent proteinases on human erythrocyte membrane skeleton proteins has been examined. Preferential cleavage of proteins 4.1 a and b and band 3 and limited cleavage of alpha- and beta-spectrin occur when either calcium-dependent proteinase I or calcium-dependent proteinase II has access to the cytoplasmic side of the ghost membrane skeleton in the presence of calcium. Thus, when these proteinases are incubated with sealed ghosts they do not cleave these proteins. Leupeptin, mersalyl, the specific cellular protein inhibitor of these enzymes, and calcium chelators can inhibit proteolysis of the red cell ghost proteins by Ca2+-dependent proteinases. Each proteinase has also been loaded into erythrocyte ghosts in the absence of calcium at low ionic strength and subsequently trapped inside by resealing the ghosts. The proteinases were activated by incubating these ghosts in the presence of the calcium ionophore A23187 and calcium. Examination of the ghost proteins by electrophoresis demonstrated calcium-dependent proteolysis of Bands 4.1 and 3 and limited cleavage of alpha- and beta-spectrin similar to that observed on proteolysis of the open, leaky ghosts. In the presence of calcium each calcium-dependent proteinase appears to associate with the erythrocyte ghost membrane.