Functional size of complement and perforin pores compared by confocal laser scanning microscopy and fluorescence microphotolysis

Confocal laser scanning microscopy and fluorescence microphotolysis (also referred to as fluorescence photobleaching recovery) were employed to study the transport of hydrophilic fluorescent tracers through complement and perforin pores. By optimizing the confocal effect it was possible to determine the exclusion limit of the pores in situ, i.e. without separation of cells and tracer solution. Single-cell flux measurements by fluorescence microphotolysis yielded information on the sample population distribution of flux rates. By these means a direct comparison of complement and perforin pores was made in sheep erythrocyte membranes. In accordance with previous studies employing a variety of different techniques complement pores were found to have a functional radius of approx. 50 A when generated at high complement concentrations. The flux rate distribution indicated that pore size heterogeneity was rather small under these conditions. Perforin pores, generated in sheep erythrocyte membranes at high perforin concentrations, were found to have a functional size very similar to complement pores. Furthermore, the functional size of the perforin pore seemed to be relatively independent of the dynamic properties of the target membrane since in two cell membranes which are very different in this regard, the human erythrocyte membrane and the plasma membrane of erythroleukemic cells, the functional radius of the perforin pore was also close to 50 A. A perforin-specific antibody reduced the functional radius of perforin pores to 45 A.     

Transients of perforin pore formation observed by fluorescence microscopic single channel recording.

A new type of single channel recording is described. Large pores were generated in the membranes of resealed human erythrocyte ghosts by incubation with perforin (cytolysin). The flux of the polar fluorescent probe Lucifer Yellow was measured in single ghosts by the fluorescence microphotolysis (photobleaching) technique. The distribution of flux rates for ghosts treated with a limiting perforin concentration showed equidistantly spaced peaks suggesting that subpopulations of ghosts with 0, 1 and 2 pores were resolved. Furthermore, distributions obtained for very different perforin concentrations could be well simulated by using one common value for the flux rate of the single pore (k = 4.65 x 10(-3) s) and assuming a Poisson distribution of pores among ghosts. The flux rate of the single pore corresponds to a pore radius of approximately 50 A, a value which is much smaller than that obtained previously by electron microscopic studies but which agrees well with recent electrical single channel recordings. Mature perforin pores were observed to be very stable. No closing events were detected at a time resolution of 0.2 s for a wide range of temperatures and Ca2+ concentrations. However, the formation of new pores was an unexpectedly slow process. Fluorescence microscopic single channel recording as introduced by this study is applicable to a variety of cellular systems and fluorescent probes and thus may complement the information obtainable by electrical single channel recording of anorganic ion fluxes.     

Cytosolic delivery of granzyme B by bacterial toxins: evidence that endosomal disruption, in addition to transmembrane pore formation, is an important function of perforin

Granule-mediated cell killing by cytotoxic lymphocytes requires the combined actions of a membranolytic protein, perforin, and granule-associated granzymes, but the mechanism by which they jointly kill cells is poorly understood. We have tested a series of membrane-disruptive agents including bacterial pore-forming toxins and hemolytic complement for their ability to replace perforin in facilitating granzyme B-mediated cell death. As with perforin, low concentrations of streptolysin O and pneumolysin (causing <10% (51)Cr release) permitted granzyme B-dependent apoptosis of Jurkat and Yac-1 cells, but staphylococcal alpha-toxin and complement were ineffective, regardless of concentration. The ensuing nuclear apoptotic damage was caspase dependent and included cleavage of poly(ADP-ribose) polymerase, suggesting a mode of action similar to that of perforin. The plasma membrane lesions formed at low dose by perforin, pneumolysin, and streptolysin did not permit diffusion of fluorescein-labeled proteins as small as 8 kDa into the cell, indicating that large membrane defects are not necessary for granzymes (32 to 65 kDa) to enter the cytosol and induce apoptosis. The endosomolytic toxin, listeriolysin O, also effected granzyme B-mediated cell death at concentrations which produced no appreciable cell membrane damage. Cells pretreated with inhibitors of endosomal trafficking such as brefeldin A took up granzyme B normally but demonstrated seriously impaired nuclear targeting of granzyme B when perforin was also added, indicating that an important role of perforin is to disrupt vesicular protein trafficking. Surprisingly, cells exposed to granzyme B with perforin concentrations that produced nearly maximal (51)Cr release (1,600 U/ml) also underwent apoptosis despite excluding a 8-kDa fluorescein-labeled protein marker. Only at concentrations of >4,000 U/ml were perforin pores demonstrably large enough to account for transmembrane diffusion of granzyme B. We conclude that pore formation may allow granzyme B direct cytosolic access only when perforin is delivered at very high concentrations, while perforin's ability to disrupt endosomal trafficking may be crucial when it is present at lower concentrations or in killing cells that efficiently repair perforin pores.     

Calreticulin, a component of the endoplasmic reticulum and of cytotoxic lymphocyte granules, regulates perforin-mediated lysis in the hemolytic model system.

Cytotoxic lymphocytes kill virally infected cells with specialized cytotoxic granules containing perforin, a protein that forms toxic pores in the target cell membrane. These specialized cytotoxic granules also contain calreticulin, an endoplasmic reticulum chaperone protein. The calcium-independent association of perforin and calreticulin prompted our evaluation of calreticulin's potential to function as a regulatory molecule that protects cytotoxic lymphocytes from their own perforin. We report here that 10(-7) M calreticulin blocked perforin-mediated lysis in the hemolytic model system using erythrocytes as targets. Previously, we found that millimolar levels of calcium in the hemolytic assays dissociate high-affinity perforin-calreticulin complexes, which makes it unlikely that perforin associates with calreticulin in solution when hemolysis is blocked. Calreticulin may affect perforin at the erythrocyte membrane. We observed calcium-dependent binding of calreticulin to erythrocyte membranes with a Kd of 2.7 x 10(-7) M and a saturation average of 10(5) molecules calreticulin per erythrocyte. At concentrations that blocked hemolysis, calreticulin occupied many of the calreticulin membrane-binding sites and was in molar excess of perforin. These observations open the possibilities that membrane-bound calreticulin prevents hydrophobic entry of perforin into membranes and (or) prevents perforin from assembling into polyperforin pores.     

Channel-forming activity of the perforin N-terminus and a putative alpha-helical region homologous with complement C9.

Cytolytic lymphocytes are endowed with a pore-forming protein called perforin. Recently, a cytolytic domain was located in the first 34 residues of the perforin N-terminus. It has been proposed that the first 19 residues are composed of a 3-domain structure including a putative amphipathic beta-sheet and that the 19 residues are sufficient for cytolytic activity. This model has now been tested by synthesizing peptides covering different portions of the N-terminus, and testing their ability to lyse lipid vesicles or increase the conductance of lipid bilayers or plasma membranes. It was found that the putative beta-sheet is indispensable for lytic activity and that the first 19 residues of the N-terminus are required for optimal lytic activity but that shorter peptides, containing only 16 residues, can form pores in lipid bilayers and cell membranes. A putative amphipathic alpha-helix from the central portion of perforin, homologous to complement C9, is nonlytic to lipid vesicles, but it can form pores in lipid bilayers. Taken together, these results support the model that the perforin N-terminus is important in initial pore formation and that the putative alpha-helical domain may be involved in subsequent perforin polymerization into large pores.     

Similarity, in molecular structure and function, between the plant toxin purothionin and the mammalian pore-forming proteins.

Many proteins containing domains of a cysteine-rich repeated motif, such as epidermal growth factor (EGF), have been reported. Here we report strong similarity between the amino acid sequence of a plant toxin--i.e., purothionin and its homologues--and with those of a domain found in mammalian pore-forming cytoplasmic proteins: components of complement and perforin of cytotoxic T-lymphocytes or natural killer-like cytotoxic cells. These similar sequences were found to be identical to the so-called EGF-like cysteine-rich repeated motif itself. Electron-microscopic observations indicated that, like complement and perforin, purothionin forms pores in the cytoplasmic membrane of target cells, resulting in their death within a few hours. On the basis of these sequence comparisons and physiological functions, we propose a scheme for the evolution of proteins containing modules of the cysteine-rich repeat motif.     

Membrane channel formation by the lymphocyte pore-forming protein: comparison between susceptible and resistant target cells

The assembly of pores by the pore-forming protein (perforin) of cytolytic T lymphocytes (CTLs) and natural killer cells on the membranes of different cell lines was studied. Using the patch clamp technique in the whole cell configuration, we measured the conductance increase induced by perforin in susceptible cell lines as well as in resistant CTL lines (CTLLs). The results showed that although the amplitudes of the first observed conductance steps produced in both cell types were comparable, CTLLs required at least 10-fold higher doses of perforin to form membrane pores. Outside-out patches excised from CTLL-R8, on the other hand, appeared to be more susceptible to channel formation by perforin than intact cells, as lower doses were able to induce conductance increases. Once channels were induced in CTL membranes, however, their conductances (greater than 1 nS) were indistinguishable from the ones obtained in susceptible cell lines. Fluorescence measurements with quin-2 showed that perforin induced rapid increases in the intracellular Ca2+ concentration in susceptible EL4 cells. In marked contrast, a perforin dose 60-120-fold higher than the minimal dose required to elicit Ca2+ changes in EL4 cells was not able to induce any measurable Ca2+ increase in CTLL-R8. The data suggest that the resistance of CTLs to lysis mediated by their own mediator perforin is at least in part due to their ability to avoid pore formation by this protein. The mechanism underlying this phenomenon is not yet understood, but the observation that outside-out patches excised from CTLL-R8 are more susceptible to channel formation by perforin than intact cells raises the possibility that an intracellular mechanism may be involved.     

Biochemical and functional characterization of a membrane-associated pore-forming protein from the pathogenic ameboflagellate Naegleria fowleri

A membrane-bound cytolytic pore-forming protein (N-PFP) produced by the pathogenic ameboflagellate Naegleria fowleri was characterized. N-PFP was solubilized from ameba membranes by detergent and enriched 300-fold by gel filtration chromatography. When analyzed by gel electrophoresis, N-PFP migrates with a molecular mass of 66 kDa and 50-54 kDa, under reducing and non-reducing conditions, respectively. In addition to lysing erythrocytes, N-PFP is cytotoxic to several tumor cell lines tested. Its hemolytic activity is not dependent on the presence of divalent cations. N-PFP rapidly depolarizes the membrane potential of microelectrode-impaled chicken embryo myocytes, suggesting that functional channel formation may represent the mode of membrane damage. In planar bilayers, N-PFP forms ion channels with heterogeneous unit conductances ranging between 150 and 400 picosiemens in 0.1 M NaCl and that are relatively resistant to closing by high voltages. Upon heat treatment (75 degrees C, 30 min), N-PFP forms channels with unit conductances that are on average larger than those formed by untreated N-PFP. N-PFP channels are slightly more permeable to cations than to anions. Using a liposome swelling-shrinkage assay, the functional diameter of N-PFP channels is estimated to range between 3.6 and 5.2 nm. N-PFP is immunologically distinct from the PFP/perforin produced by lymphocytes, the terminal components of complement and a PFP from the ameba Entamoeba histolytica, all of which produce pores on target membranes. This protein may have a direct lytic role during target cell killing mediated by N. fowleri.     

Localization of erythrocyte membrane antigens by immune electron microscopy

Torus (T) protein was prepared by differential centrifugation of hemoglobin-free membranes after freezing, thawing and dialysis against water. Ferritin-labeled antibody to T protein failed to agglutinate erythrocytes or to sensitize them to complement lysis. Anti-T bound specifically to the internal aspect of ghosts and to the exterior of inverted vesicles prepared from them. The findings lend further support to the supposition that native T protein exists only on the inner aspect of intact erythrocyte membranes; its possible relation to spectrin is uncertain. In contrast, ferritin-labeled antibody to purified erythrocyte glycoprotein (virus receptor substance, VRS) agglutinated erythrocytes, sensitized them to complement lysis and by its specific reactivity with surface antigen afforded a basis for the electron microscopic quantitation of receptor sites on plasma membranes examined in thin section as well as with freeze-etching.     

Permeability characteristics of cell-membrane pores induced by ostreolysin A/pleurotolysin B,binary pore-forming proteins from the oyster mushroom

Proteins from the oyster mushroom, 15 kDa ostreolysin A (OlyA), and 59 kDa pleurotolysin B (PlyB) with a membrane attack complex/perforin (MACPF) domain, damage cell membranes as a binary cytolytic pore-forming complex. Measurements of single-channel conductance and transmembrane macroscopic current reveal that OlyA/PlyB form non-selective ion-conducting pores with broad, skewed conductance distributions in N18 neuroblastoma and CHO-K1 cell membranes. Polyethylene-glycol 8000 (hydrodynamic radius of 3.78 nm) provides almost complete osmotic protection against haemolysis, which strongly suggests a colloid-osmotic type of erythrocyte lysis. Our data indicate that OlyA/PlyB form transmembrane pores of varied sizes, as other pore-forming proteins with a MACPF domain.     

Temperature effects on osmotic fragility, and the erythrocyte membrane

Results are reported on the temperature-dependence of intact-cell surface area, isotonic volume, hemolytic volume, and ghost steady-state surface area and volume, using several techniques of resistive pulse spectroscopy. Temperature was found not to alter the intact cell surface area permanently: the area remains constant at 130 +/- 1 micron 2, at temperatures ranging from 0 to 40 degrees C. Temperature does alter the steady-state volume of the cells, with a colder temperature inducing swelling by about 0.29 micron 3/deg. C. Such a temperature-induced volume change is sufficient to explain only approximately half of the fragility differences which result from temperature changes. The remainder was found to result from higher temperatures enabling a substantial transient increase in surface area of intact cells (up to at least 14% of 40 degrees C), with a corresponding increase in the cell's hemolytic volume (up to 21%). The hemolytic volume apparently increases linearly with temperature, since steady-state ghost volumes are found to increase linearly with the temperature at which the ghosts were produced. In the steady state (at high temperature), the membranes of electrically-impermeable resealed ghosts can remain extended by more than 10%, compared with membranes of the corresponding unhemolyzed, intact red cells.     

Functional reconstitution of influenza virus envelopes.

We have examined several procedures for the reconstitution of influenza virus envelopes, based on detergent removal from solubilized viral membranes. With octylglucoside, no functionally active virosomes are formed, irrespective of the rate of detergent removal: in the final preparation the viral spike proteins appear predominantly as rosettes. Protein incorporation in reconstituted vesicles is improved when a method based on reverse-phase evaporation of octylglucoside-solubilized viral membranes in an ether/water system is employed. However, the resulting vesicles do not fuse with biological membranes, but exhibit only a non-physiological fusion reaction with negatively charged liposomes. Functional reconstitution of viral envelopes is achieved after solubilization with octaethyleneglycol mono(n-dodecyl)ether (C12E8), and subsequent detergent removal with Bio-Beads SM-2. The spike protein molecules are quantitatively incorporated in a single population of virosomes of uniform buoyant density and appear on both sides of the membrane. The virosomes display hemagglutination activity and a strictly pH-dependent hemolytic activity. The virosomes fuse with erythrocyte ghosts, as revealed by a fluorescence resonance energy transfer assay. The rate and the pH dependence of fusion are essentially the same as those of the intact virus. The virosomes also fuse with cultured cells, either at the level of the endosomal membrane or directly with the cellular plasma membrane upon a brief exposure to low pH.     

Calorimetric study of interactions of toxin from Radianthus macrodactylus with erythrocyte membrane.

Differential scanning microcalorimetry, intrinsic protein fluorescence and SDS-electrophoresis have been applied for the study of the hemolytic toxin from sea anemone, Radianthus macrodactylus, when it is incorporated into pure sphingomyelin/water systems and upon its effect on the human and dog erythrocyte membranes. The results obtained by using these techniques showed that one molecule of toxin withdraws six sphingomyelin molecules from the cooperative transition. Effect of hemolytic toxin on human erythrocyte ghosts causes an appearance of a new heat sorption peak in their differential scanning calorimetric curve with a maximum of 36 degrees C and, moreover, it leads to a loss of one of cytoskeleton proteins (actin). These effects are essentially weaker in the case of dog erythrocyte ghosts. This suggests differences in the structural organization of human and dog erythrocyte membranes.     

Fusion of membrane vesicles bearing only the influenza hemagglutinin with erythrocytes, living cultured cells, and liposomes

Membrane vesicles, bearing only the influenza viral hemagglutinin glycoprotein, were reconstituted following solubilization of intact virions with Triton X-100. The viral hemagglutinin glycoprotein was separated from the neuraminidase glycoprotein by agarose sulfanilic acid column. The hemagglutinin glycoprotein obtained was homogenous in gel electrophoresis and devoid of any neuraminidase activity. A quantitative determination revealed that the hemolytic activity of the hemagglutinin vesicles was comparable to that of intact virions. Incubation of fluorescently labeled hemagglutinin vesicles with human erythrocyte ghosts (HEG) or with liposomes composed of phosphatidylcholine/cholesterol or phosphatidylcholine/cholesterol/gangliosides, at pH 5.0 but not at pH 7.4, resulted in fluorescence dequenching. Very little, if any, fluorescence dequenching was observed upon incubation of fluorescently labeled HA vesicles with neuraminidase or glutaraldehyde-treated HEG or with liposomes composed only of phosphatidylcholine. Hemagglutinin vesicles were rendered non-hemolytic by treatment with NH2OH or glutaraldehyde or by incubation at 85 degrees C or low pH. No fluorescence dequenching was observed following incubation of non-hemolytic hemagglutinin vesicles with HEG or liposomes. These results clearly suggest that the fluorescence dequenching observed is due to fusion between the hemagglutinin vesicles and the recipient membranes. Incubation of hemagglutinin vesicles with living cultured cells, i.e. mouse lymphoma S-49 cells, at pH 5.0 as well as at pH 7.4, also resulted in fluorescence dequenching. The fluorescence dequenching observed at pH 7.4 was inhibited by lysosomotropic agents (methylamine and ammonium chloride) as well as by EDTA and NaN3, indicating that it is due to fusion of hemagglutinin vesicles taken into the cells by endocytosis.     

Analysis of Perforin Assembly by Quartz Crystal Microbalance Reveals a Role for Cholesterol and Calcium-independent Membrane Binding

Perforin is an essential component in the cytotoxic lymphocyte-mediated cell death pathway. The traditional view holds that perforin monomers assemble into pores in the target cell membrane via a calcium-dependent process and facilitate translocation of cytotoxic proteases into the cytoplasm to induce apoptosis. Although many studies have examined the structure and role of perforin, the mechanics of pore assembly and granzyme delivery remain unclear. Here we have employed quartz crystal microbalance with dissipation monitoring (QCM-D) to investigate binding and assembly of perforin on lipid membranes, and show that perforin monomers bind to the membrane in a cooperative manner. We also found that cholesterol influences perforin binding and activity on intact cells and model membranes. Finally, contrary to current thinking, perforin efficiently binds membranes in the absence of calcium. When calcium is added to perforin already on the membrane, the QCM-D response changes significantly, indicating that perforin becomes membranolytic only after calcium binding.     

Capturing pore-forming intermediates of MACPF and binary toxin assemblies by cryoEM

Deployed by both pathogenic bacteria and host immune systems, pore-forming proteins rupture target membranes and can serve as conduits for effector proteins. Understanding how these proteins work relies on capturing assembly intermediates. Advances in cryoEM allowing in silico purification of heterogeneous assemblies has led to new insights into two main classes of pore-forming proteins: membrane attack complex perforin (MACPF) proteins and binary toxins. The structure of an immune activation complex, sMAC, shows how pores form by sequential templating and insertion of β-hairpins. CryoEM structures of bacterial binary toxins present a series of transitions along the pore formation pathway and reveal a general mechanism of effector protein translocation. Future developments in time-resolved cryoEM could capture and place short-lived states along the trajectory of pore-formation.     

Identification and partial characterization of a novel plasma membrane-associated lytic factor isolated from highly purified adherent lymphokine-activated killer cells.

We have identified and partially purified a novel cytolytic factor isolated from enriched plasma membranes prepared from highly purified lymphokine-activated killer cells (adherent-LAK. A-LAK cells) and a large granular lymphocytic NK cell leukemia, CRNK-16. The enriched plasma membranes were shown to be physically devoid of lytic granules and contained no detectable pore-forming protein (PFP, perforin) activity. The plasma membrane-associated cytolytic factor (designated M-CTX) was solubilized in biologically active form and was highly lytic to a large panel of target cells in 2- to 4-hr 51Cr release assays. Characteristics of the M-CTX include: (1) it is plasma membrane- not granule-associated: (2) it is not hemolytic and functions in the absence of Ca2+: (3) nucleated target cells are lysed in 2 to 4 hr at 37 degrees C but not at 4 degrees C: (4) it induces apoptotic cell death with nuclear DNA fragmentation and massive membrane blebbing: (5) it is isolated from the plasma membranes of cultured A-LAK cells, a lytically active LGL leukemia (CRNK-16), and fresh spleen cells but not from thymocytes or L929 fibroblasts: and (6) the lytic activity of the partially purified toxin is inactivated by trypsin, serum, and heat, but is not blocked by antibodies that inactivate TNF-alpha, LT or IFN-gamma. Taken collectively, these data suggest that M-CTX may represent a heretofore undescribed membrane-associated toxin possibly involved in contact-mediated cell killing.     

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.     

P-glycoprotein does not protect cells against cytolysis induced by pore-forming proteins

P-glycoprotein (P-gp) is an ATP-dependent drug pump that confers multidrug resistance (MDR). In addition to its ability to efflux toxins, P-gp can also inhibit apoptosis induced by a wide array of cell death stimuli that rely on activation of intracellular caspases for full function. We therefore hypothesized that P-gp may have additional functions in addition to its role in effluxing xenotoxins that could provide protection to tumor cells against a host response. There have been a number of contradictory reports concerning the role of P-gp in regulating complement activation. Given the disparate results obtained by different laboratories and our published results demonstrating that P-gp does not affect cell death induced by another membranolytic protein, perforin, we decided to assess the role of P-gp in regulating cell lysis induced by a number of different pore-forming proteins. Testing a variety of different P-gp-expressing MDR cell lines produced following exposure of cells to chemotherapeutic agents or by retroviral gene transduction in the complete absence of any drug selection, we found no difference in sensitivity of P-gp(+ve) or P-gp(-ve) cells to the pore-forming proteins complement, perforin, or pneumolysin. Based on these results, we conclude that P-gp does not affect cell lysis induced by pore-forming proteins.