Staphylococcal alpha-toxin: oligomerization of hydrophilic monomers to form amphiphilic hexamers induced through contact with deoxycholate detergent micelles

Staphylococcal alpha-toxin increases the permeability of lipid vesicles by cholesterol- and pH-dependent assembly of oligomeric channels

alpha-Toxin, a lethal hemolytic toxin secreted by Staphylococcus aureus, forms ionic channels of large size in lipid membranes. To investigate the mechanism of channel assembly we have studied the kinetics of pore formation on small unilamellar vesicles. We have used two assays of vesicle permeabilization: one is the release of a fluorescent molecule trapped in their inner compartment; the other is the dissipation of an imposed potential. Both methods indicate that the kinetics are complex consisting of an initial delay followed by a non-linear relaxation. The dependence of the pore formation rate and the extent of permeabilization on the toxin/vesicle ratio indicates that aggregation of 4-10 preinserted toxin monomers underlies channel assembly. The pH dependence of permeabilization suggests that protonation of an acidic group of the toxin is a prerequisite to channel formation. Inclusion of cholesterol in the target vesicles potentiates alpha-toxin effects, in a dose-dependent way, possibly by facilitating its protonation. The location of the proton-binding site on the two adjacent aspartic acid residues in positions 127 and 128 of the toxin monomer is proposed.

Potent leukocidal action of Escherichia coli hemolysin mediated by permeabilization of target cell membranes

The contribution of Escherichia coli hemolysin (ECH) to bacterial virulence has been considered mainly in context with its hemolytic properties. We here report that this prevalent bacterial cytolysin is the most potent leukocidin known to date. Very low concentrations (approximately 1 ng/ml) of ECH evoke membrane permeability defects in PMN (2-10 x 10(6) cells/ml) leading to an efflux of cellular ATP and influx of propidium iodide. The attacked cells do not appear to repair the membrane lesions. Human serum albumin, high density and low density lipoprotein, and IgG together protect erythrocytes and platelets against attack by even high doses (5-25 micrograms/ml) of ECH. In contrast, PMN are still permeabilized by ECH at low doses (50-250 ng/ml) in the presence of these plasma inactivators. Thus, PMN become preferred targets for attack by ECH in human blood and protein-rich body fluids. Kinetic studies demonstrate that membrane permeabilization is a rapid process, ATP-release commencing within seconds after application of toxin to leukocytes. It is estimated that membrane permeabilization ensues upon binding of approximately 300 molecules ECH/PMN. This process is paralleled by granule exocytosis, and by loss of phagocytic killing capacity of the cells. The recognition that ECH directly counteracts a major immune defence mechanism of the human organism through its attack on granulocytes under physiological conditions sheds new light on its possible role and potential importance as a virulence factor of E. coli.

Quantitative study of the binding and hemolytic efficiency of Escherichia coli hemolysin

Mono- and polyclonal antibodies were used to construct a sandwich enzyme-linked immunosorbent assay that permitted quantitation of Escherichia coli hemolysin in soluble and membrane-bound forms. Toxin concentrations of 4 to 14 micrograms/ml were measured in culture supernatants of E. coli LE 2001 at times of peak hemolytic activity. Quantitative studies on the binding of E. coli hemolysin to rabbit erythrocytes were conducted at 0 and 37 degrees C. At 37 degrees C, 85 to 95% of bindable toxin was cell bound after 60 min, and no saturability of binding was observed in the studied range of concentrations, which resulted in deposition of approximately 100 to 50,000 toxin molecules per cell. Binding was slower and less effective at 0 degrees C; however, hemolysis did occur at low temperature. The number of cell-bound toxin molecules required to generate a hemolytic lesion within 60 min was estimated to be approximately 100 molecules per cell at 37 degrees C and 800 to 1,000 molecules per cell at 0 degrees C. Upon prolonged incubation (5 to 20 h, 37 degrees C), the number of molecules evoking a functional lesion decreased to approximately 5 to 20 per cell. These results are compatible with the concept that E. coli hemolysin first adsorbs to the cell surface, with membrane insertion and pore formation following in a second step that may be temporally dissociated from that of binding. The data support the pore concept of toxin action by showing that attachment of a low and finite number of toxin molecules to an erythrocyte will ultimately generate a cytolytic lesion.

Cytolysis by Ca-permeable transmembrane channels. Pore formation causes extensive DNA degradation and cell lysis

This study investigates the effect of the purified membrane pore formers, staphylococcal alpha-toxin and CTL perforin, on target cell lysis as measured by 51Cr release and on nuclear damage as measured by DNA degradation and 125IUdR release. Both pore formers cause dose-dependent cell lysis, which is accompanied by DNA release. The ratio of DNA/Cr release depends on the nature of target cell and shows the same pattern as the ratio of release of the two markers reported for CTL-mediated lysis of the same targets. DNA degradation is dependent on the presence of intracellular Ca in the target cell and is totally blocked if Ca is chelated by Quin 2 intracellularly and EGTA extracellularly. DNA degradation, in addition, is inhibited by the lysosomotropic agents NH4Cl, chloroquine, and monensin. rTNF doubles the degree of DNA degradation mediated by alpha-toxin in 3-h assays. We conclude that pore formers alone can mediate DNA degradation. In addition, they may promote the uptake of other factors and thereby accelerate their time course of action. DNA degradation by pore formers requires active target participation in a pathway that is dependent on intracellular Ca and lysosomes. These aspects of target lysis resemble CTL- and NK cell-mediated cytolysis.

Incorporation of staphylococcal alpha-toxin in glutaraldehyde fixed erythrocytes

The incorporation of staphylococcal alpha-toxin into glutaraldehyde fixed erythrocytes occurs in the same way as with native erythrocytes. Binding of alpha-toxin to the cells is accompanied by oligomerization of native 3 S toxin to the membrane-bound 11 S toxin hexamer, which is embedded into the lipid bilayer of the membrane. Antibodies against alpha-toxin, build up during an infection with S. aureus, can be determined in a passive hemagglutination test (IHT) using glutaraldehyde fixed and alpha-toxin treated erythrocytes. To test the validity of this IHT, antibodies to alpha-toxin were determined in 550 human sera of patients from hospitals of the University of Giessen suspected to suffer from staphylococcal infections and in 300 sera of healthy blood donors. The results were compared with the titres obtained by a convenient neutralisation test (ASTA). All sera with elevated titres in the ASTA test also showed high titres in the IHT. Because it is simple to perform and highly reproducible, the IHT seems to be a valuable test for detection of antibodies against staphylococcal alpha-toxin.

Effect of staphylococcal alpha-toxin on intracellular Ca2+ in polymorphonuclear leukocytes

Staphylococcal alpha-toxin, a channel-forming protein, stimulates leukotriene B4 formation in rabbit polymorphonuclear leukocytes (PMN) (N. Suttorp, W. Seeger, J. Zucker-Reimann, L. Roka, and S. Bhakdi, Infect. Immun. 55:104-110, 1987). The concept was advanced that transmembrane toxin pores act as Ca2+ gates allowing passive Ca2+ influx into the cell, thus initiating stimulus response coupling. A critical step in this hypothesis is the demonstration of an increase in the cytosolic free Ca2+ concentration [( Ca2+]i). [Ca2+]i and membrane-associated Ca2+ were therefore monitored in quin-2- or chlorotetracycline-loaded PMN exposed to alpha-toxin. The effects of the Ca2+ ionophore ionomycin and the chemotactic tripeptide formylmethionyl-leucylphenylalanine (fMLP) were studied in parallel. All stimuli increased [Ca2+]i in dose- and time-dependent manner. In the presence of an EDTA excess there was a decrease of [Ca2+]i due to an efflux of Ca2+ in alpha-toxin- and ionomycin-treated cells, while addition of fMLP still induced an increase of [Ca2+]i. In the presence of verapamil, a Ca2+ channel blocker, [Ca2+]i was reduced after stimulation with fMLP but not with alpha-toxin or ionomycin. Addition of fMLP and ionomycin but not of alpha-toxin to PMN resulted in a rapid and substantial mobilization of membrane-associated Ca2+. The collective data demonstrate that exposure of PMN to staphylococcal alpha-toxin results in an increase in [Ca2+]i which is due to an influx of extracellular Ca2+ and not to a mobilization of intracellularly stored Ca2+. The concept of initiating stimulus response coupling by Ca2+ influx through transmembrane pores may be generally applicable to other channel-forming cytolysins.

Bacterial exotoxins and endothelial permeability for water and albumin in vitro

Effects of Staphylococcus aureus alpha-toxin and Pseudomonas aeruginosa cytotoxin on the permeability of an endothelial monolayer were studied. Porcine pulmonary artery endothelial cells were grown on a polycarbonate membrane, mounted in a chamber, and exposed to a continuous hydrostatic pressure of 10 cmH2O. On application of this trans-endothelial pressure, endothelial monolayer became "sealed," i.e., the filtration rate for water decreased and the reflection coefficient for albumin increased, reaching a plateau after 1-2 h. Sealed monolayer had a hydraulic conductivity of 2.1 X 10(-6) cm.s-1.cmH2O and an albumin reflection coefficient of 0.73. Permeability of the monolayer was increased on addition of an excess of EDTA and reversed on readdition of calcium. Within 60-90 min after addition of 1 microgram/ml alpha-toxin, the filtration rate increased 75-fold, and the albumin reflection coefficient dropped to 0.20. These changes in permeability were accompanied by cell retraction and formation of large intercellular gaps between endothelial cells. Effects of alpha-toxin were abolished by preincubation with neutralizing antibodies and by inhibitors of calmodulin function. Pseudomonas aeruginosa cytotoxin (25 and 50 micrograms/ml) also increased the permeability of the endothelial monolayer, but it was only about one-third as effective as alpha-toxin.

Inactivation of a T cell receptor-associated GTP-binding protein by antibody-induced modulation of the T cell receptor/CD3 complex

TCR modulation induced by anti-TCR or anti-CD3 mAbs leads to a transient state of refractoriness of the T cell to all signals given via cell surface structures. To investigate the underlying mechanisms, we have used human CTL permeabilized with the alpha toxin of S. aureus. This method of permeabilization allows manipulation of the interior milieu of the cell, but maintains its functional and structural integrity. Introduction of the G protein activator GTP gamma S into permeabilized CTL leads to triggering of granule exocytosis. The G protein inactivator GDP beta S inhibited exocytosis induced by TCR triggering but not that induced by activation of protein kinase C. This indicates that the G protein that triggers exocytosis is localized after CD3 triggering but before formation of the polyphosphoinositol breakdown product diacylglycerol. In TCR-modulated CTL, GTP gamma S is no longer able to activate exocytosis. Such CTL, however, still respond to PKC activators. This demonstrates that a TCR-associated G protein has been functionally inactivated by TCR modulation.

Staphylococcal alpha toxin promotes blood coagulation via attack on human platelets

Staphylococcus aureus plays a major role as a bacterial pathogen in human medicine, causing diseases that range from superficial skin and wound to systemic nosocomial infections . The majority of S. aureus strains produces a toxin, a proteinaceous exotoxin whose hemolytic, dermonecrotic, and lethal properties have long been known (1-6). The toxin is secreted as a single- chained, nonglycosylated polypeptide with a M(r) of 3.4 x 10(4) (7, 8). The protein spontaneously binds to lipid monolayers and bilayers (9-14), producing functional transmembrane pores that have been sized to 1.5-2.0-nm diameters (15-18). The majority of pores formed at high toxin concentrations (20 mug/ml) is visible in the electron microscope as circularized rings with central pores of approximately 2 nm in diameter. The rings have been isolated, and molecular weight determinations indicate that they represent hexamers of the native toxin (7). We have proposed that transmembrane leakiness is due to embedment of these ring structures in the bilayer, with molecular flux occurring through the central channels (15, 19). Pore formation is dissectable into two steps (20, 21). Toxin monomers first bind to the bilayer without invoking bilayer leakiness . Membrane-bound monomers then laterally diffuse and associate to form non-covalently bonded oligomers that generate the pores. When toxin pores form in membranes of nucleated cells, they may elicit detrimental secondary effects by serving as nonphysiologic calcium channels, influx of this cation triggering diverse reactions, including release of potent lipid mediators originating from the arachidonate cascade (22-24). That alpha toxin represents an important factor of staphylococcal pathogenicity has been clearly established in several models of animal infections through the use of genetically engineered bacterial strains deleted of an active alpha toxin gene (25-27). Whether the toxin is pathogenetically relevant in human disease, however, is a matter of continuing debate. Doubts surrounding this issue originate from two main findings. First, whereas 60 percent hemolysis of washed rabbit erythrocytes is effected by approximately 75 ng/ml alpha toxin, approximately 100-fold concentrations are required to effect similar lysis of human cells (4-6, 13). The general consensus is that human cells display a natural resistance towards toxin attack. The reason for the wide inter-species variations in susceptibility towards alpha toxin is unknown but does not seem to be due to the presence or absence of high-affinity binding sites on the respective target cells (20, 21). Second, low-density lipoprotein (28) and neutralizing antibodies present in plasma of all healthy human individuals inactivate a substantial fraction of alpha toxin in vitro. These inactivating mechanisms presumably further raise the concentration threshold required for effective toxin attack, and it is most unlikely that such high toxin levels will ever be encountered during infections in the human organism. The aforegoing arguments rest on the validity of two general assumptions. First, the noted natural resistance of human erythrocytes to alpha toxin must be exhibited by other human cells. Second, toxin neutralization by plasma components, usually tested and quantified after their preincubation with toxin in vitro, must be similarly effective under natural conditions, and protection afforded by these components must not be restricted to specific cell species.

The cytolytic toxin aerolysin must aggregate to disrupt erythrocytes, and aggregation is stimulated by human glycophorin

The hole-forming toxin aerolysin was shown to aggregate after binding to erythrocytes at 37 degrees C. Although the protein also bound and aggregated at 4 degrees C, hole formation was not observed, indicating that aggregation preceded penetration of the lipid bilayer. Aggregation, but not binding, could be blocked by pretreatment of the toxin with diethyl pyrocarbonate, a histidine-reactive reagent. This resulted in inactivation of the toxin. Incubation of aerolysin with glycophorin purified from human erythrocytes caused aggregation and complete inactivation. Erythrocytes which lacked glycophorin were less sensitive to the toxin. Proaerolysin, the inactive precursor of aerolysin, also bound to erythrocytes; however, it did not aggregate, nor did it aggregate when preincubated with glycophorin. The protoxin could be activated by treatment with trypsin even after it had bound to erythrocytes. Activation could also be achieved by reaction of proaerolysin with a variety of other proteases, each of which brought about a similar reduction in protein molecular weight. The activated protein was resistant to further proteolysis. These results indicate that aggregation is a necessary step in hole formation and that the sites on aerolysin required for binding and for aggregation and hole formation are separate.

Aggregation of IgE receptors induces degranulation in rat basophilic leukemia cells permeabilized with alpha-toxin from Staphylococcus aureus

A method has been developed by which rat basophilic leukemia (RBL) cells can be permeabilized to small molecules while maintaining their ability to degranulate in response to aggregation of IgE receptors. alpha-Toxin was isolated from culture supernatants of Staphylococcus aureus by precipitation with (NH4)2SO4 and chromatography on phenyl-Sepharose. The isolated toxin binds to the plasma membrane of RBL cells and polymerizes to form a transmembrane pore that allows small molecules (Mr less than 1000), but not macromolecules, to diffuse freely across the membrane. There was no spontaneous release of the contents of RBL cell secretory granules during permeabilization or subsequent incubations. Substantial IgE receptor-mediated exocytosis occurred in the absence of Ca2+, and degranulation was maximal at 0.1-1.0 microM Ca2+, the physiologically important range of Ca2+ concentrations. Using these permeabilized cells, small molecules (i.e., substrates and inhibitors of various enzymes) normally excluded by the plasma membrane can be introduced into the cell. Moreover, the intracellular concentration of ions (such as Ca2+) can be precisely controlled. This method will allow a detailed examination of the individual biochemical events involved in degranulation of mast cells.

Hydrodynamic properties of colicin A. Existence of a high-affinity lipid-binding site and oligomerization at acid pH

The hydrodynamic properties of colicin A have been studied. The molecular mass of colicin A was determined from sedimentation equilibrium centrifugation to be 63 +/- 1.2 kDa, in agreement with that determined from the primary amino acid sequence [Morlon et al. (1983) J. Mol. Biol. 110, 271-289]. The sedimentation coefficient has been analyzed over a wide range of ionic strength (NaCl 0.06-0.56 M) and pH (8-4) and was found to remain almost constant. However, below pH 5 an oligomerization of colicin A to tetramers occurred. The frictional coefficient value indicated that the shape of the colicin A monomer was very asymmetric. Analysis of the pH dependence of circular dichroism of colicin A and of its COOH-terminal domain indicated that a sharp transition occurred between pH 4 and 3. This transition was very much reduced for the COOH-terminal domain in the presence of a non-ionic detergent. The presence of a lipid-binding site in colicin A at neutral pH was demonstrated both by hydrodynamic studies with micelles of n-hexadecanoyl and n-octadecanoylphosphocholine and by differential sensitivity to a proteolytic enzyme in the presence or absence of detergent micelles. About 75 molecules of lipid were bound under these conditions suggesting that colicin A was bound to lipid micelles. In contrast, at acid pH, in the presence of an excess of lipid the tetramer was dissociated into monomers complexed to 20-30 lipid molecules, indicating the exposure of a high-affinity lipid-binding site.

Quantitative analysis of the binding and oligomerization of staphylococcal alpha-toxin in target erythrocyte membranes

The binding of staphylococcal alpha-toxin to rabbit and human erythrocytes was quantitated over a wide range of toxin concentrations (3 x 10(-11) to 3 x 10(-6) M) with the use of an enzyme-linked immunosorbent assay that permitted simultaneous quantitation of monomeric and oligomeric toxin forms. Three basic observations were made. First, in no range of concentrations did the binding of alpha-toxin to rabbit erythrocytes display characteristics of a receptor-ligand interaction. Net binding to rabbit cells was nil at sublytic concentrations (10(-10) M or 3 ng/ml). The onset of binding occurred at around 10 ng/ml and remained fairly constant and ineffective (5 to 8% of toxin offered) over a wide concentration range (up to 10 micrograms/ml). Second, hemolysis of rabbit and human erythrocytes at 37 degrees C was always accompanied by the formation of toxin oligomers in the membrane. Third, overall toxin binding at 0 degree C followed a pattern similar to that at 37 degrees C. However, oligomer formation and cell lysis were retarded (but not totally inhibited) at 0 degree C. When rabbit erythrocytes were incubated with low levels of toxin at 0 degree C (0.5 microgram/ml) for 30 min, the toxin became bound exclusively in monomer form, and no lysis occurred. When cells thus treated were washed and suspended at 37 degrees C, lysis rapidly ensued, and native monomeric toxin was replaced by oligomeric toxin. The collective results directly support the oligomer pore concept of toxin action and also indicate that toxin oligomers form by lateral aggregation of bound monomers in the bilayer. They speak against the existence of specific binding sites for alpha-toxin on rabbit erythrocytes.

Quantitation of monomeric and oligomeric forms of membrane-bound staphylococcal alpha-toxin by enzyme-linked immunosorbent assay with a neutralizing monoclonal antibody

A murine monoclonal antibody generated against staphylococcal alpha-toxin was shown to react only with the monomeric (native), 3S form of the toxin. A sensitive sandwich enzyme-linked immunosorbent assay (ELISA) constructed with this antibody permitted detection of 0.25 to 0.5 ng of native toxin per ml. Toxin oligomers formed either by heat aggregation in solution, on target erythrocyte membranes, or on phosphatidylcholine-cholesterol liposomes were unreactive in the ELISA when membranes were solubilized with the nondenaturing detergent Triton X-100. After dissociation of the oligomers by boiling in sodium dodecyl sulfate, however, the ELISA reactivity of the liberated 3S toxin was fully restored. Parallel determinations of membrane-bound toxin with sodium dodecyl sulfate and Triton X-100 solubilization thus permitted direct quantitation of total and monomeric toxin, respectively; the difference between these two values was represented by toxin oligomers. The detection limits for membrane-bound oligomeric and monomeric toxin on erythrocyte membranes are in the order of 100 molecules and 1 molecule per cell, respectively. Using this ELISA, we show that over 90% of alpha-toxin molecules bound to target membranes at 37 degrees C are in oligomeric form. Evidence is given that the monoclonal antibody neutralizes alpha-toxin by inhibiting its binding to both rabbit and human erythrocytes. This ELISA is the first assay that quantitatively discriminates between mono- and oligomeric forms of a pore-forming protein on target cell membranes.

Further characterization of dopamine release by permeabilized PC12 cells

Rat pheochromocytoma cells (PC12) permeabilized with staphylococcal alpha-toxin release [3H]dopamine after addition of micromolar Ca2+. This does not require additional Mg2+-ATP (in contrast to bovine adrenal medullary chromaffin cells). We also observed Ca2+-dependent [3H]-dopamine release from digitonin-permeabilized PC12 cells. Permeabilization with alpha-toxin or digitonin and stimulation of the cells were done consecutively to wash out endogenous Mg2+-ATP. During permeabilization, ATP was removed effectively from the cytoplasm by both agents but the cells released [3H]dopamine in response to micromolar Ca2+ alone. Replacement by chloride of glutamate, which could sustain mitochondrial ATP production in permeabilized cells, does not significantly alter catecholamine release induced by Ca2+. However, Mg2+ without ATP augments the Ca2+-induced release. The release was unaltered by thiol-, hydroxyl-, or calmodulin-interfering substances. Thus Mg2+-ATP, calmodulin, or proteins containing -SH or -OH groups are not necessary for exocytosis in permeabilized PC12 cells.

Conformation of colicin A: apparent difference between cytoplasmic and extracellular polypeptide chain

Cytoplasmic colicin A has the ability to bind to membranes and to form stable dimers. This form remains stable even in the presence of 1% SDS at 25 degrees C. Both of these properties were not observed for extracellular colicin A suggesting a possible difference in the conformation between cytoplasmic and extracellular colicin A.

Cell membrane interaction of Bacillus thuringiensis subsp. israelensis cytolytic toxins

Two toxic polypeptides of 24 and 25 kilodaltons (kDa) were purified from parasporal proteinaceous crystals of Bacillus thuringiensis subsp. israelensis. Both of these polypeptides, which are antigenically similar and have identical N terminals, lysed human erythrocytes and cultured mosquito cells. Although the 24-kDa peptide was more toxic than the 25-kDa peptide, both were less toxic than the crude alkali-solubilized crystal toxin. However, a 1:1 mixture of these 24- and 25-kDa proteins was more toxic than either of these polypeptides individually, indicating a possible interaction between these proteins at the cell membrane. Both the 24- and the 25-kDa proteins were inactivated by aqueous suspensions of dioleolylphosphatidylcholine, indicating the involvement of phospholipids in the cytotoxic action of these toxins. Thus the role of cell membrane phospholipids in mediating the toxin action was studied by using phospholipases as probes. Treatment of erythrocytes with high levels of phospholipase D increased their susceptibility to the toxin; however, phospholipase A2-treated erythrocytes were less susceptible to the toxin. These erythrocytes also bound less 125I-labeled 25-kDa toxin. These results support the role of fatty acyl residues at the syn-2 position of membrane phospholipids in toxin action. The cytolytic toxin of B. thuringiensis subsp. israelensis is thought to damage cell membranes in a detergentlike manner. However, there was a difference between the cytolytic action of this toxin and that of a nonionic detergent such as Triton X-100 because phospholipase A2-treated erythrocytes were more susceptible to Triton X-100, whereas such erythrocytes were less sensitive to the toxin. Thus, the cytolytic toxin apparently did not act as a nonspecific detergent, but rather interacted with phospholipid receptors on the cell membrane. Such an interaction of the toxin with phospholipid receptors probably results in the increased cell permeability, thereby causing cell lysis.

Pore formation by Staphylococcus aureus alpha-toxin in lipid bilayers. Dependence upon temperature and toxin concentration

Staphylococcus aureus alpha-toxin forms ionic channels of large size in lipid bilayer membranes. We have developed two methods for studying the mechanism of pore formation. One is based on measurement of the ionic current flowing through a planar lipid membrane after exposure to the toxin; the other is based on measuring the release of the fluorescent complex Tb-Dipicolinic acid from large unilamellar vesicles under similar conditions. Both methods indicate that the pore formation process is complex, showing an initial delay followed by non-linear kinetics. The power dependence of the pore formation rate on the toxin concentration in planar bilayers indicates that an aggregation mechanism underlies the channel assembly. Arrhenius plots, obtained with both techniques, show no deviation from linearity up to 50 degrees C and the derived activation energies are found to be comparable to those for the binding and the lysis of rabbit erythrocytes by the same toxin. The temperature dependence of the conductance induced in planar bilayers by a large number of toxin channels indicates that the pores are filled with aqueous solution. The analysis of single conductance events shows that a heterogeneous population of pores exist and that smaller channels are preferred at low temperature. We attribute this heterogeneity to the existence of pores resulting from the aggregation of different numbers of monomers.

Mechanism of leukotriene generation in polymorphonuclear leukocytes by staphylococcal alpha-toxin

The effects of staphylococcal alpha-toxin on arachidonic acid metabolism in rabbit polymorphonuclear leukocytes (PMNs) were investigated and compared with those of the ionophore A23187 and the chemotactic tripeptide formylmethionyl-leucyl-phenylalanine (fMLP). Sublytic amounts of alpha-toxin stimulated the release of leukotriene B4 (LTB4) in PMNs in a dose-dependent manner. The toxin was several times more potent than fMLP but was not as effective as the ionophore. Preincubation of the toxin with neutralizing antibodies abolished the effect. Extracellular calcium was strictly required for eliciting LTB4 generation. Verapamil, a calcium channel blocker, inhibited fMLP-mediated LTB4 generation but had no effect on alpha-toxin- or A23187-exposed PMNs. Agents such as trifluoperazine and N-6(aminohexyl)-5-chloro-1-naphthalene sulfonamid that interfered with calmodulin activity, however, inhibited LTB4 generation in all cases. One minute after the addition of alpha-toxin, PMNs exhibited a severalfold enhancement in passive permeability to 45Ca2+. In addition, these cells became permeable to sucrose but not to inulin or dextran. The influx pattern was consistent with the previous observation that alpha-toxin creates discrete transmembrane channels in erythrocytes with an effective internal diameter of 2 to 3 nm. The results suggest that alpha-toxin triggers the arachidonic acid pathway in PMNs by facilitating calcium influx into the cells, possibly via transmembrane toxin pores that serve as calcium gates. Generation of arachidonic acid metabolites in PMNs by sublytic amounts of alpha-toxin may represent an important cellular reaction that generally occurs during infections with Staphylococcus aureus.

Use of a monoclonal antibody to determine the mode of transmembrane pore formation by streptolysin O

Murine monoclonal antibodies were generated against streptolysin O. One out of 10 tested immunoglobulin clones exhibited strong neutralizing activity; in solution, the presence of approximately two to four antibody molecules per toxin monomer effected 50% neutralization of hemolytic toxin activity. An enzyme-linked immunosorbent assay performed with target cell membranes that were treated with streptolysin O in the presence and absence of neutralizing antibodies showed that the antibodies did not block primary binding of the toxin to the cells. When membranes were solubilized in deoxycholate detergent and centrifuged in sucrose density gradients, those lysed with streptolysin O contained detergent-resistant, high-molecular-weight oligomers identical to the pore lesions, whereas those given toxin and neutralizing antibody contained the toxin exclusively in low-molecular-weight, nonoligomerized form. The process of pore formation by streptolysin O must thus involve two distinct steps, i.e., the primary binding of toxin molecules to the membrane followed by oligomerization of bound toxin monomers by lateral aggregation in the lipid bilayer to form the transmembrane pores.

Ionic channels formed by Staphylococcus aureus alpha-toxin: voltage-dependent inhibition by divalent and trivalent cations

The interaction of Staphylococcus aureus alpha-toxin with planar lipid membranes results in the formation of ionic channels whose conductance can be directly measured in voltage-clamp experiments. Single-channel conductance depends linearly on the solution conductivity suggesting that the pores are filled with aqueous solution; a rough diameter of 11.4 +/- 0.4 A can be estimated for the pore. The conductance depends asymmetrically on voltage and it is slightly anion selective at pH 7.0, which implies that the channels are asymmetrically oriented into the bilayer and that ion motion is restricted at least in a region of the pore. The pores are usually open in a KCl solution but undergo a dose- and voltage-dependent inactivation in the presence of di- and trivalent cations, which is mediated by open-closed fluctuations at the single-channel level. Hill plots indicate that each channel can bind two to three inactivating cations. The inhibiting efficiency follows the sequence Zn2+ greater than Tb3+ greater than Ca2+ greater than Mg2+ greater than Ba2+, suggesting that carboxyl groups of the protein may be involved in the binding step. A voltage-gated inactivation mechanism is proposed which involves the binding of two polyvalent cations to the channel, one in the open and one in the closed configuration, and which can explain voltage, dose and time dependence of the inactivation.

Activation of cytolytic T lymphocyte and natural killer cell function through the T11 sheep erythrocyte binding protein

The T11 sheep erythrocyte binding glycoprotein [relative molecular mass (Mr)50,000(50K)] is expressed throughout human T-lymphocyte ontogeny and appears to play an important physiological role in T-cell activation. Thus, the treatment of T cells with certain monoclonal anti-T11 antibodies results in antigen-independent polyclonal T-cell activation as assessed by proliferation and lymphokine secretion. In addition, the majority of thymocytes that have not yet acquired the T3-Ti antigen/major histocompatibility complex (MHC) receptor can be activated to express interleukin-2 (IL-2) receptors through this T11 structure. We show here that the triggering of cytolytic T (Tc) cells via T11 causes an antigen-independent activation of the cytolytic mechanism as evidenced by the induction of nonspecific cytolytic activity. Furthermore, T11+T3-Ti- natural killer (NK) cell clones can also be induced to lyse NK-cell-resistant targets by treatment with anti-T11 monoclonal antibodies directed at defined T11 epitopes. These results indicate that T11 triggering can activate cytotoxic lymphocytes to express their functional programmes in the absence of specific antigen recognition via the T3-Ti complex and provide further evidence for the notion that certain NK cells and T lymphocytes are related.

Secondary structure and assembly mechanism of an oligomeric channel protein

The alpha-toxin of Staphylococcus aureus is secreted as a water-soluble, monomeric polypeptide (Mr 33 182) that can assemble into an oligomeric membrane channel. By chemical cross-linking, we have confirmed that the major form of the channel is a hexamer. The circular dichroism spectrum of this hexamer in detergent revealed that it contains a high proportion of beta-sheet that we deduce must lie within the lipid bilayer when the protein is associated with membranes. The circular dichroism spectrum of the monomeric toxin in the presence or absence of detergent was closely similar to the spectrum of the hexamer, suggesting that the secondary structure of the polypeptide is little changed on assembly. Results of experiments involving limited proteolysis of the monomer and hexamer are consistent with the idea that assembly involves the movement of two rigid domains about a hinge located near the midpoint of the polypeptide chain. The hydrophilic monomer is thereby converted to an amphipathic rod that becomes a subunit of the hexamer.

Pseudomonas aeruginosa cytotoxin stimulates prostacyclin production in cultured pulmonary artery endothelial cells: membrane attack and calcium influx

The effects of highly purified Pseudomonas aeruginosa cytotoxin were investigated on cultured pulmonary artery endothelial cells. This toxin dose-dependently (7.5-60 micrograms/ml) and time-dependently (20-75 minutes) stimulated the release of radiolabeled arachidonic acid and metabolites and the synthesis of prostacyclin in the absence of overt cell damage (no enhanced lactate dehydrogenase [LDH] release). Preincubation of the toxin with neutralizing antibodies abolished the effect. The toxin response on endothelial cells required extracellular calcium but not magnesium and was accompanied by a calcium influx. Interference with intracellular calcium function by TMB 8 or with (calcium)-calmodulin function by trifluoperazine and W7 dose-dependently reduced the cytotoxin mediated synthesis of prostacyclin. Calcium channel blockers (nimodipine, diltiazem, verapamil, D 888), however, were ineffective in this system. Following addition of cytotoxin to endothelial cells, an increased passive permeability for small marker molecules (potassium, 45calcium, 3H-sucrose), but for large ones (3H-inulin, 3H-dextran, LDH) was noted, suggesting that cytotoxin creates discrete hydrophilic transmembrane lesions of about 0.5-1.5 nm in diameter. These data are compatible with the notion that Pseudomonas aeruginosa cytotoxin triggers the arachidonic acid pathway in cultured pulmonary artery endothelial cells by calcium influx and suggest that this calcium influx may proceed through toxin created transmembrane lesions.

Purification of oligomeric staphylococcal alpha-toxin by affinity chromatography on digitonin-sepharose

An effective concentration of alpha-toxin from Staphylococcus aureus Wood 46, directly from the culture supernatant, could be achieved by adsorption on digitonin-sepharose and elution with 3 mol/l sodium thiocyanate (NaSCN). The toxin was further purified by gelchromatography. The purified product yielded 1 single protein band upon SDS-polyacrylamide electrophoresis. It was nonhemolytic, but reacted with anti-alpha-toxin under complement fixation. Dialysis against 0.14 mol/l NaCl with hydrophobic amino acids partially reactivated the alpha-hemolytic activity of the toxin. Ultracentrifugal analysis yielded sedimentation coefficients for the purified toxin of approximately 3,7 S when dissolved in 3 mol/l NaSCN and of about 12 S after dialysis against 0.14 mol/l NaCl (Table 1). The spontaneous oligomerization of the alpha-toxin during dialysis against 0.14 mol/l NaCl possibly resulted from a change in configuration induced by its adsorption to digitonin-sepharose.

Staphylococcal alpha-toxin-induced PGI2 production in endothelial cells: role of calcium

Studies in erythrocytes indicate that staphylococcal alpha-toxin generates discrete transmembrane channels with an effective diameter of 2-3 nm. In cultured, confluent, pig pulmonary arterial endothelial cells we studied the triggering of the arachidonic acid cascade and its dependence on calcium influx, possibly through toxin-created pores. In endothelial cells alpha-toxin time dependently (5-30 min) and dose dependently (0.1-8 micrograms/ml) stimulated the release of radiolabeled arachidonic acid and prostacyclin (PGI2) production in similar amounts as the calcium ionophore A23187 (10 microM). Preincubation of alpha-toxin with neutralizing antibodies abolished the effect. The toxin response was strictly dose dependent on extracellular calcium but not on magnesium. The toxin effect was accompanied by an up to 10-fold increased passive permeability of pulmonary arterial endothelial cells for 45Ca. Interference with calcium-calmodulin function (trifluoperazine, W7) dose dependently reduced production of PGI2, but blockers of physiological calcium channels (verapamil, nimodipine, nisoldipine, and diltiazem) did not. In contrast to the effect of the ionophore A23187, the toxin effect was accompanied by a release of potassium, but in neither system was there a release of lactate dehydrogenase. In addition, alpha-toxin but not ionophore-exposed endothelial cells showed an increased passive influx of small radiolabeled markers (45Ca and [3H]sucrose) but not of large markers [( 3H]inulin and [3H]dextran). These data are consistent with the concept that alpha-toxin triggers the arachidonic acid cascade in pulmonary arterial endothelial cells by calcium influx and suggest that this calcium influx may proceed through toxin-created transmembrane channels.

Correlation between toxin binding and hemolytic activity in membrane damage by staphylococcal alpha-toxin

The binding of Staphylococcus aureus alpha-toxin to rabbit and human erythrocytes was studied by hemolytic assays and sodium dodecyl sulfate-polyacrylamide gel electrophoresis immunoblotting. Hemolytic assays showed that toxin binding to 10% cell suspensions at neutral pH was very ineffective in the concentration range 3 X 10(-8) to 3 X 10(-7) M (1 to 10 micrograms/ml), and less than 5% of added toxin became cell bound. However, binding was augmented as toxin levels were raised, abruptly increasing to 50 to 60% at 2 X 10(-6) to 3 X 10(-6) M (60 to 100 micrograms/ml). When rabbit erythrocytes were lysed with 1 to 5 micrograms of toxin per ml, both monomeric and hexameric forms of the toxin could be detected on the membranes by sodium dodecyl sulfate-polyacrylamide gel electrophoresis immunoblotting. In contrast, human erythrocytes treated with 1 to 6 micrograms of toxin per ml did not lyse, and membrane-bound toxin was not detectable. When toxin concentrations were raised to 30 to 100 micrograms/ml, human erythrocytes also lysed and toxin hexamers became membrane bound in comparable amounts as on rabbit cell membranes. Lowering the pH led to a marked increase in susceptibility of human, but not rabbit erythrocytes towards alpha-toxin. When human cells were lysed at pH 5.0 with 5 micrograms of toxin per ml, membrane-bound hexameric toxin became detectable. The demonstrated correlation between the presence of hexameric, cell-bound toxin and hemolytic activity supports the channel concept of toxin-mediated cytolysis. The results also show that toxin binding does not exhibit overall characteristics of a simple receptor-ligand interaction.

Mechanism of complement cytolysis and the concept of channel-forming proteins

Complement damages membranes via the terminal reaction sequence that leads to the formation of membrane-bound, macromolecular C5b-9(m) protein complexes. These complexes represent C5b-8 monomers to which varying numbers of C9 molecules can be bound. Complexes carrying high numbers of C9 (ca. 6/8-12/16?) exhibit the morphology of hollow protein channels. Because they are embedded within the lipid bilayer, aqueous transmembrane pores are generated that represent the primary lesions caused by complement in the target cell membrane. Many other proteins damage membranes by forming channels in a manner analogous to the C5b-9(m) complex. Two prototypes of bacterial exotoxins, Staphylococcus aureus alpha-toxin and streptolysin-O, are discussed in this context, and attention is drawn to the numerous analogies existing among these protein systems. Common to all is the process of self-association of the native proteins to form supramolecular complexes. This event is in turn accompanied by a unique transition of the molecules from a hydrophilic to an amphiphilic state.

Staphylococcal alpha-toxin elicits hypertension in isolated rabbit lungs. Evidence for thromboxane formation and the role of extracellular calcium

Staphylococcal alpha-toxin is known to damage mammalian cell membranes. Studies of erythrocytes indicate that the native toxin generates a discrete transmembrane channel with an effective diameter of 2-3 nm. (Füssle, R., S. Bhakdi, A. Szeigoleit, J. Tranum-Jensen, T. Kranz, and H.J. Wellensiek. 1981. J. Cell Biol. 91:83-94.) In isolated rabbit lungs, perfused with recirculating blood- and plasma-free perfusion fluid, the mediation of a toxin-provoked vascular pressor response by the triggering of the arachidonic acid cascade and its dependence on extracellular calcium were investigated. Dose-dependent pulmonary artery pressor responses were elicited by the injection of 0.5-5 micrograms staphylococcal alpha-toxin into the pulmonary artery. The pressor responses were completely abolished by preincubation of the toxin with neutralizing antibodies or by preformation of alpha-toxin hexamers in vitro. They were accompanied by the release of the arachidonic acid metabolites thromboxane B2 and 6-keto-prostaglandin F1 alpha (stable metabolites of thromboxane A2 and prostaglandin I2, respectively) into the perfusion fluid. They were blocked by inhibitors of thromboxane synthetase, cyclooxygenase, and phospholipase, as well as by substances that interfere with calcium-calmodulin function. alpha-Toxin induced selective release of potassium, but not lactatedehydrogenase into the medium. Calcium depletion of the intravascular space did not suppress the toxin-dependent potassium release but did abrogate the pressor response and the release of the arachidonic acid metabolites. When calcium was reintroduced into the circulation without the application of a second toxin stimulus, marked pressor responses paralleled by the release of arachidonic acid metabolites occurred. The conclusion drawn from these studies is that staphylococcal alpha-toxin provokes pulmonary vascular hypertension which is apparently mediated by thromboxane A2 formation, which surpasses the biological effect of the simultaneously formed prostaglandin I2. The triggering of the arachidonic acid cascade is strictly dependent on extracellular calcium and may be mediated by a nonphysiological calcium bypass through transmembrane toxin channels with subsequent stimulation of phospholipase activity.

Generation of leukotrienes from human granulocytes by alveolysin from Bacillus alvei

We investigated the effect of alveolysin on human granulocytes. Alveolysin is an exoprotein produced by Bacillus alvei and belongs to the group of sulfhydryl-activated cytolysins. Other members of this group are streptolysin O and theta-toxin from Clostridium perfringens. It is demonstrated that alveolysin leads to leukotriene generation from human granulocytes, which exert chemotactic (leukotriene B4) and slow-reacting substance (leukotriene C4, D4, and E4) activity under sublytic concentrations.

Chromatofocusing: a new method for purification of staphylococcal enterotoxins B and C1

A new chromatographic procedure was developed which obtained highly purified preparations of staphylococcal enterotoxins B and C1 in yields of 60% from cultures of Staphylococcus aureus and which is faster than any of the separation methods used previously. The procedure involves chromatography on carboxymethylcellulose, removal of alpha-toxin by adsorption to rabbit erythrocyte membranes, and finally, chromatofocusing as the fundamental new step. Enterotoxins were obtained in highly purified form and behaved in a homogeneous manner as determined by ultracentrifugation and electrophoresis on polyacrylamide gel in the presence of sodium dodecyl sulfate, with molecular weights of 34,000 for staphylococcal enterotoxin B and 30,000 for staphylococcal enterotoxin C1. Using chromatofocusing as the final purification step, we isolated three B and six C1 distinct but immunologically identical enterotoxin fractions, which were found to be devoid of any impurities and to possess a marked degree of toxicity in monkeys.

Cytolysis by H-2-specific T killer cells. Assembly of tubular complexes on target membranes

Cloned T killer cells derived from one-way mixed lymphocyte reactions were characterized with regard to their Lyt phenotype and specificity. Two clones of Lyt-1-2+ phenotype that recognized H-2Dd were selected and examined for their cytolytic function by negative staining and thin section electron microscopy. When incubated with the H-2d target S194, they assemble two types of tubular complexes, polyperforin 1 and 2. Both structures appear to arise by polymerization of precursors that may originate in dense granules and/or Golgi of T killer cells. Polyperforins appear to be associated with vesicles that are released during the lytic reaction and transferred to target membranes as shown by immunoelectron microscopy. Since there is a close correlation between target lysis and the appearance of polyperforins on target membranes, it is suggested that polyperforins take part in the cell-mediated killing reaction. Although polyperforins are different in size and several molecular properties from complement, there are striking similarities between these circular complexes and polyperforin (C9). It is therefore possible that they belong to a closely related family of cytolytic effector molecules.

Assembly of two types of tubules with putative cytolytic function by cloned natural killer cells

The formation of ultrastructural membrane lesions of varying size during cell mediated cytolysis effected by human peripheral blood leukocytes was recently reported by Dourmashkin et al. and Henkart et al. Using cloned mouse natural killer (NK) cells as effectors and YAC-1 cells or rabbit erythrocytes as targets, we now report two types of membrane lesions with inner diameters of 16 +/- 2 nm and approximately 5 nm, respectively. These lesions arise by membrane insertion of tubular complexes that may be assembled from subunits during the cytolytic reaction. The tubules are detected on target membranes by immune electron microscopy and appear to form transmembrane channels as seen in ultrathin sections. Both tubules are partially purified by membrane extraction with SDS and gel filtration in deoxycholate containing buffer. Based on the correlation of tubule assembly and cytolysis and on their detection on target membranes, we suggest that both types of tubules may be related to cytolysis.