Morphological changes in Ehrlich ascites tumor cells during the cell fusion reaction with HVJ (Sendai virus): II. Cluster formation of intramembrane particles in the early stage of cell fusion

The morphological events in the cell membrane of Ehrlich ascites tumor (EAT) cells associated with cell fusion caused by HVJ were investigated with freeze-fracture technique. When cell fusion was carried out at 37 °C, the EATC fusion was too rapid to allow identification of the sequential steps of membrane fusion and no deleterious changes in the plasma membrane could be detected. However, on lowering the incubation temperature from 37 to 28 °C, the process of cell fusion was slower and there was a distinct alteration in the plasma membrane. On incubation of cell aggregates with HVJ at 28 °C, the fusion reaction proceeded very slowly. On incubation for 10 min, fusion was initiated in a few cells, but most of the cells remained agglutinated with their cell membranes close to those of neighboring cells and often in direct contact in small localized regions. When cells in this stage were chilled and fixed at 4 °C, large clusters of intramembrane particles (IMPs) were seen all over the P face. On further incubation of the cells at 37 °C, cell fusion proceeded rapidly and the IMPs became randomly redistributed, indicating that clustering is a reversible phenomenon occurring in the early stage of cell fusion. This clustering was temperature-dependent. It was seen in cell fixations at 4 °C, but not at 28 °C without chilling, and it was prevented by inhibitors of cell fusion, such as cytochalasin D (CD) or glucose at high concentration. These findings suggest that certain structural changes in the plasma membrane that may induce thermotropic aggregation of IMP are required to initiate cell fusion.     

Modification of cell membranes with viral envelopes during fusion of cells with HVJ (Sendai virus).

A large number of viral materials are associated with the surface of cells after cell fusion with HVJ at 37 °C for 30 min. This is due to fusion of viral envelopes with the cell membrane. Studies were made on the process from viral adsorption to cell-cell, or cell-viral envelope fusion. On incubation at low temperatures, such as 0–15 °C, no envelope fusion or cell fusion was observed, although there was some interaction between the virus and cells. This interaction resulted in loss of hemadsorption (HA) activity of the cells and partial damage of the ion barrier of the cell membrane. The viral particles seem to come close to the lipid layer of the cell membrane at the low temperatures and to distort the non-flexible membrane structure. On incubation of the cell-virus complex at 37 °C, the cells rapidly became HA-positive and the HA activity was maximal within 5 min. At this stage there was much leakage of ions through the cell membrane. On further incubation the damage to the ion barrier of the cell membrane was repaired completely with completion of cell fusion. This process may be correlated with fusion of viral envelopes with cell membranes and restoration of the cell membrane fused with them.     

Effects of cytochalasin D on fusion of cells by HVJ (Sendai virus).

Fusion of cells mediated by HVJ was inhibited completely with 5 μg/ml or more of cytochalasin D (CD). With cytochalasin, HVJ-cell interaction at 0 °C proceeded as well as without cytochalasin; HVJ was adsorbed to cell surfaces and the cells agglutinated together. Then the virus particles were enfolded with cell membranes, which resulted in the disappearance of hemadsorption activity on the cell surfaces. When the cell-virus complex was incubated at 37 °C, the early reactions proceeded as well as without cytochalasin; the hemadsorption activity reappeared on the cell surfaces, the viral envelopes fused with cell membranes at the same degree as without cytochalasin, and a stage sensitive to sodium azide appeared as in a control without cytochalasin. But cell-to-cell fusion did not occur in the presence of cytochalasin; cells were dissociated freely from the cell aggregates during incubation. This indicates that cell-to-cell fusion was inhibited but HVJ envelope to cell membrane interactions proceeded well on incubation at 37 °C. These findings suggest that viral envelope-cell membrane fusion and cell-cell fusion are separable, and participation of a cytoskeleton system including microfilaments in the cells is essential for cell-cell fusion.     

Modification of cell membranes with viral envelopes during fusion of cells with HVJ (Sendai virus) : I. Interaction between cell membranes and virus in the early stage

A large number of viral materials are associated with the surface of cells after cell fusion with HVJ at 37 °C for 30 min. This is due to fusion of viral envelopes with the cell membrane. Studies were made on the process from viral adsorption to cell-cell, or cell-viral envelope fusion. On incubation at low temperatures, such as 0–15 °C, no envelope fusion or cell fusion was observed, although there was some interaction between the virus and cells. This interaction resulted in loss of hemadsorption (HA) activity of the cells and partial damage of the ion barrier of the cell membrane. The viral particles seem to come close to the lipid layer of the cell membrane at the low temperatures and to distort the non-flexible membrane structure. On incubation of the cell-virus complex at 37 °C, the cells rapidly became HA-positive and the HA activity was maximal within 5 min. At this stage there was much leakage of ions through the cell membrane. On further incubation the damage to the ion barrier of the cell membrane was repaired completely with completion of cell fusion. This process may be correlated with fusion of viral envelopes with cell membranes and restoration of the cell membrane fused with them.     

Mouse C2 myoblast cells resist HVJ (Sendai virus)-mediated cell fusion in the proliferating stage but become capable of fusion after differentiation

To investigate the mechanism of myoblast fusion, we attempted to prepare artificial myotubes of mouse C2 myoblast cells using the hemagglutinating virus of Japan (HVJ, Sendai virus). Proliferating C2 cells showed strong resistance to HVJ-mediated cell fusion and remained morphologically unchanged even though massive numbers of virions adsorbed onto their surface. They showed no membrane disruption, which occurs in the early stage of cell fusion induced by HVJ. These observations suggest that proliferating C2 cells are resistant to HVJ-mediated cell fusion. However, upon induction of differentiation, C2 cells gradually became capable of fusion induced by HVJ and then even generated heterokaryons with Ehrlich ascites tumor cells. When differentiated C2 cells that had become fusion-sensitive were treated with HVJ in the presence of EDTA, they did not fuse but degenerated, suggesting that their cell membranes were transiently disrupted by interaction with HVJ. These results suggest that the cell membranes of myoblasts change to a fusion-capable state during the process of differentiation.     

Modification of cell membranes with viral envelopes during fusion of cells with HVJ (Sendai virus) : II. Effects of pretreatment with a small number of HVJ

Ehrlich ascites tumor cell membranes were completely modified after incubation at 37 °C for 30 min with a small dose of HVJ (about 0.7% of the maximum number of the virus particles that could be adsorbed onto the cells). After this treatment, the cells could adsorb further added HVJ onto their surfaces at 0 °C. But the cell agglutination which was induced by viral adsorption at 0 °C was very weak, and the interaction of the adsorbed virus with the lipid layer of the cell membrane at 37 °C preceding fusion or lysis of the cells was not strong. A discrepancy was observed between acquisition of the modification and liberation of sialic acid (destruction of viral receptors) by viral neuraminidase. The modification proceeded well on incubation at 37 °C but not at lower temperatures. The possibility that the modification is induced by fusion of viral envelopes with cell membranes is discussed.     

Characterisation of myogenic cell membrane: II. Dynamic changes in membrane lipids during the differentiation of mouse C2 myoblast cells

Proliferating mouse C2 myoblast cells resist haemagglutinating virus of Japan, Sendai virus (HVJ) mediated cell fusion. However, differentiating C2 cells can be induced to fuse by HVJ, suggesting that the rigid membrane of C2 cells changes during the differentiation. To investigate this phenomenon, changes in membrane lipids which affect fluidity were examined. Membrane cholesterol gradually decreased with the differentiation of C2 cells. However, spontaneous fusion to form myotubes and artificial fusion induced by HVJ were both inhibited when the level of cholesterol was prevented from falling in the cell membrane. The membranes of differentiating C2 cells contained more unsaturated fatty acids than those of proliferating cells. Thus, when differentiating C2 cells were treated with stearate (a saturated fatty acid), they failed to form myotubes and were insensitive to HVJ-mediated fusion. Whereas, if proliferating C2 cells were given linolenate (an unsaturated fatty acid), they became capable of HVJ-induced fusion. These results indicate that differentiating C2 cells change their fusion sensitivity by decreasing cholesterol, probably at the same time as they increase the unsaturated fatty acid content of the cell membrane.     

Morphological changes in Ehrlich ascites tumor cells during the cell fusion reaction with HVJ (Sendai virus). III. Morphological characterization of HVJ glycoproteins integrated into the plasma membrane and their internalization by coated vesicles

On cell-cell fusion of Ehrlich ascites tumor (EAT) cells with HVJ, HVJ envelopes also fuse with the cell membrane, resulting in integration of the viral envelope glycoproteins into the fused cell membranes. Morphological characterization of the glycoproteins in the plasma membrane and the mode of their internalization were investigated in detail. In the fusion reaction, the glycoproteins were rapidly integrated into the cell membrane within 2 or 3 min on incubation at 37 °C and they remained at the fusion sites, not dispersing widely, during further incubation. Thus they were still present in clusters in the plasma membrane at the end of the fusion reaction. On culture of fused cells in culture medium, internalization of the viral glycoproteins was initiated by formation of coated vesicles and most of the integrated glycoproteins were endocytosed into the cytoplasm within 30 min. Soon after internalization, the coated vesicles fused with each other, losing their coat materials. The intact virions that remained unfused on the cell surface were also internalized, but coat materials did not appear on the inside surface of the cell membrane, unlike in the case of integrated glycoproteins.     

Fusion of Madin-Darby canine kidney cells by HVJ (Sendai virus): absence of direct association of virus particles with the site of membrane fusion

The process of cell fusion of Madin-Darby canine kidney (MDCK) cells by HVJ (Sendai virus) was investigated to determine whether the HVJ particles were directly associated with the site of membrane fusion. Confluent monolayer cultures of MDCK cells are sealed together by tight junctions on the apices of their lateral membranes, so added virus particles can be adsorbed only to the apical surfaces of the cells. After incubation with HVJ at 37 degrees C for 30 min, the cells still appeared mononucleate and unfused by light microscopy, but electron microscopic examination showed that fusion at the lateral membranes had occurred below the tight junctions. Furthermore, when fluorescein isothiocyanate (FITC)-labeled macromolecules, which cannot pass across the gap junctions, were injected into the cells at this stage, labeled macromolecules were found to diffuse into the adjacent cells. These findings strongly suggest that cell fusion was initiated in the lateral membrane, a region distinct from the site of adsorbed HVJ particles. Thus, the virus particles were not directly associated with the fusion site, but induced fusion of the lateral membranes indirectly.     

Pathway of vesicular stomatitis virus entry leading to infection

The entry of vesicular stomatitis virus into Madin-Darby canine kidney (MDCK) cells was examined both biochemically and morphologically. At low multiplicity and 0 °C, viruses bound to the cell surface but were not internalized. Binding was very dependent on pH. More than ten times more virus bound at pH 6.5 than at higher pH values. At the optimal pH, binding failed to reach equilibrium after more than two hours. The proportion of virus bound was irreproducible and low, relative to the binding of other enveloped viruses. Over 90% of the bound viruses were removed by proteases. When cells with pre-bound virus were warmed to 37 °C, a proportion of the bound virus became protease-resistant with a half-time of about 30 minutes. After a brief lag period, degraded viral material was released into the medium. The protease-resistant virus was capable of infecting the cells and probably did so by an intracellular route, since ammonium chloride blocked the infection and slightly reduced the degradation of viral protein.When the entry process was observed by electron microscopy, viruses were seen bound to the cell surface at 0 °C and, after warming at 37 °C, within coated pits, coated vesicles and larger, smooth-surfaced vesicles. No fusion of the virus with the plasma membrane was observed at pH 7.4.When pre-bound virus was incubated at a pH below 6 for 30 seconds at 37 °C, about 40 to 50% of the pre-bound virus became protease-resistant. On the basis of this result and previously published experiments (White et al., 1981), it was concluded that vesicular stomatitis virus fuses to the MDCK cell plasma membrane at low pH.These experiments suggest that vesicular stomatitis virus enters MDCK cells by endocytosis in coated pits and coated vesicles, and is transported to the lysosome where the low pH triggers a fusion reaction ultimately leading to the transfer of the genome into the cytoplasm. The entry pathway of vesicular stomatitis virus thus resembles that described earlier for both Semliki Forest virus and fowl plague virus.     

Virus Meets Liposome

Abstract

Sendai virus was the first virus to encounter liposomes. Gangliosides when incorporated into liposomes act as Sendai virus receptors even at 0–4°C. When receptor-containing liposomes are incubated with virus at 37°C, they envelop the virus. At 37°C liposomes also fuse with Sendai virus membrane.

Virus binding initially involves weak adhesion, which may allow the virus to “browse” the cell, and which is followed by adhesion strengthening. MicrogrΔpHs of Sendai virus fusion with liposomes after one minute at 37°C indicate that fusion occurs at the very curved leading edge of the region of the liposome enveloping virus. A model of fusion is proposed that emphasizes the role of the curvature and membrane tension in this localized region of “host” membrane. The curvature assists close approach and destabilizes the outer monolayer. The proposed intermediates are consistent with the “stalk” hypothesis.     

Electron microscopic study on the interaction of Sendai virus with liposomes containing glycophorin

The interaction of liposomes containing glycophorin, a major sialoglycoprotein of human crythrocytes, with Sendai virus was studied by freeze-fracture and negative staining electron-microscopy. Viral envelopes were absorbed on liposomal membranes at 0°C. When the temperature was shifted up to 37°C, the viral envelopes fused with the liposomal membranes (envelope fusion). Particles representing viral membrane components formed clusters on liposomal membranes after incubation for more than 1 h at 37°C.     

A spin-label study on fusion of red blood cells induced by hemagglutinating virus of Japan.

Fusion of red blood cells (RBC) induced by hemagglutinating virus of Japan (HVJ) has been studied using a phosphatidylcholine spin label. The spin label was readily incorporated and diffused into the lipid bilayer portion of the viral envelope. The exchange broadening in the electron spin resonance (ESR) spectrum of densely labeled virus disappeared rapidly when the virus was mixed with RBC at 37 degrees. The spectrum gradually approached that of the host cell spin labeled with the phosphatidylcholine label. The results directly indicate transfer and intermixing of phospholipid molecules between the viral envelope and RBC membrane. The transfer reaction was strongly dependent on temperature. No transfer was observed at lower temperatures where the virus adsorbed to the cell and caused aggregation but no hemolysis and fusion. The transfer rate remained negligibly small until 19 degrees and increased rapidly between 25 and 30 degrees. The virus-induced hemolysis showed similar temperature dependence. The transfer rate was greatly reduced under inhibitory conditions of fusion: glutaraldehyde treatment of RBC, trypsin treatment of HVJ, or the presence of concanavalin A. Only slight transfer was observed from fusion-inactive influenza virus to RBC. The transfer was greatly enhanced by the help of HVJ. The close parallelism suggests that the transfer and intermixing are necessary steps to the cell fusion. The transfer rate was dependent on fluidity of the host cell membrane and independent of the viral dose. The virus-induced transfer of phospholipid molecules between RBC's was also detected by the spin label. Its temperature dependence was quite similar to that for the virus-to-cell transfer. The intercellular transfer was nearly proportional to the viral dose.     

Influence of the spectrin network on fusion of influenza virus with red blood cells

We examined the influence of the physical state of the membrane skeleton on low pH fusion of influenza virus A/PR 8/34 with intact human red blood cells. Spectrin, the major component of the skeleton, is known to become denaturated at 50°C. After heat treatment of erythrocytes at 50°C we observed an enhanced kinetics of fusion monitored spectrofluorometrically by the octadecylrhodamine fluorescence dequenching assay, while the extent of fusion was not affected. The accelerated fusion of influenza virus after preincubation of red blood cells at 50°C is not mediated by alterations of the lipid phase of the target. From ESR measurements using spin-labelled phospholipids we conclude that heat-induced alterations of the spectrin network did not affect either the phospholipid asymmetry or the fluidity of the exoplasmic and the cytoplasmic leaflets of the erythrocyte membrane. Moreover, as deduced from our previous investigations, the swelling behaviour of red blood cells could not be responsible for the observed effect. Possible mechanisms for the spectrin effect include a change in the ability of the target membrane to bend locally, and a change in the rate of formation and development of the fusion pore.     

Efficient introduction of contents of liposomes into cells using HVJ (Sendai virus)

The conditions for efficient introduction of the contents of liposomes into cells were examined using fragment A of diphtheria toxin (DA) as a marker; one molecule of DA can kill a cell when introduced into the cytoplasm. Liposomes containing DA (DA liposomes) were toxic to cells treated with HVJ (Sendai virus) at 4 degrees C just before exposure to DA liposomes at 37 degrees C, but were not toxic to untreated cells. This toxicity was temperature-dependent. DA outside of liposomes was not toxic to HVJ-treated cells. Results also showed that liposomes could fuse with HVJ at 37 degrees but not at 4 degrees C and that liposomes preincubated with HVJ at 37 degrees C could associate with cells. DA liposomes preincubated with HVJ at 37 degrees C were highly toxic to cells. This toxicity was dependent on the duration of preincubation with HVJ and the dose of HVJ. When plasmid DNA coded herpes simplex virus thymidine kinase was trapped in liposomes and fused with Ltk- cells with HVJ, the thymidine kinase activity was expressed in about 10% of the cells. These data show that naked liposomes fuse efficiently with cells with HVJ and that the contents of the liposomes can be introduced into the cytoplasm 100-10 000 times more efficiently by treatment of the cells or liposomes with HVJ.     

Sendai virus-erythrocyte membrane interaction: quantitative and kinetic analysis of viral binding, dissociation, and fusion.

A kinetic and quantitative analysis of the binding and fusion of Sendai virus with erythrocyte membranes was performed by using a membrane fusion assay based on the relief of fluorescence self-quenching. At 37 degrees C, the process of virus association displayed a half time of 2.5 min; at 4 degrees C, the half time was 3.0 min. The fraction of the viral dose which became cell associated was independent of the incubation temperature and increased with increasing target membrane concentration. On the average, one erythrocyte ghost can accommodate ca. 1,200 Sendai virus particles. The stability of viral attachment was sensitive to a shift in temperature: a fraction of the virions (ca. 30%), attached at 4 degrees C, rapidly (half time, ca. 2.5 min) eluted from the cell surface at 37 degrees C, irrespective of the presence of free virus in the medium. The elution can be attributed to a spontaneous, temperature-induced release, rather than to viral neuraminidase activity. Competition experiments with nonlabeled virus revealed that viruses destined to fuse do not exchange with free particles in the medium but rather bind in a rapid and irreversible manner. The fusion rate of Sendai virus was affected by the density of the virus particles on the cell surface and became restrained when more than 170 virus particles were attached per ghost. In principle, all virus particles added displayed fusion activity. However, at high virus-to-ghost ratios, only a fraction actually fused, indicating that a limited number of fusion sites exist on the erythrocyte membrane. We estimate that ca. 180 virus particles maximally can fuse with one erythrocyte ghost.     

Functional modification of Sendai virus by siRNA

Sendai virus (hemagglutinating virus of Japan; HVJ) is a negative-strand RNA virus with robust fusion activity, and has been utilized for gene transfer and drug delivery. Hemagglutinin-neuraminidase (HN) protein on the viral membrane is important for cell fusion, but causes agglutination of red blood cells. HN-depleted HVJ has been desired for in vivo transfection in order to improve safety. Here, we succeeded in producing HN-depleted HVJ using HN-specific short interfering RNA (siRNA). Viral production was not affected by the siRNA. HN protein was markedly decreased in the new HVJ, while other viral proteins were retained. Consequently, the hemagglutinating activity was substantially reduced and infection activity was suppressed. When the HN-depleted HVJ was mixed with cultured cells and the mixture was centrifuged for 10min at 2000xg, the modified HVJ recovered its infectivity to approximately 80% of wild HVJ. However, infectivity was abolished in the presence of anti-F antibody. Moreover, transfection of FITC-labeled oligodeoxynucleotides using the modified HVJ was also recovered by centrifugation. Thus, the HN-depleted HVJ produced using siRNA technology will be applicable to a delivery vector.     

Kinetic analysis of fusion of hemagglutinating virus of Japan with erythrocyte membrane using spin-labeled phosphatidylcholine

HVJ* (hemagglutinating virus of Japan containing spin-labeled phosphatidylcholine in its envelope around 10 mol %) was adsorbed onto erythrocytes or erythrocyte ghosts at various doses, and the ESR spectrum of the virus-cell system was measured at 37 degrees C. The peak-height increase for the HVJ*-ghost system was satisfactorily analyzed on the basis of envelope fusion by a first-order kinetic equation with two different rate constants. The rate constant was obtained as k1 = 0.84 min-1 and k2 = 0.011 min-1, independent of the virus dose. The fraction of virus fused at the rate constant k1 decreased with the dose. However, the average number of fast-fusing viruses per cell was nearly independent of the dose, and the value was one to two. The peak-height increase in the HVJ*-erythrocyte system was caused by both envelope fusion and phospholipid exchange catalyzed by the virus-induced hemolyzate. At lower doses, where the virus-induced hemolysis was small and, therefore, the rate of phospholipid exchange was small, the peak-height increase could be analyzed by the same kinetic equation with nearly the same rate constant value for k1 as that for HVJ*-ghosts. However, the k2 was larger than that for HVJ*-ghost, owing to the additional transfer by phospholipid exchange.     

The role of stearate attachment to the hemagglutinin‐esterase‐fusion glycoprotein HEF of influenza C virus

The only spike of influenza C virus, the hemagglutinin‐esterase‐fusion glycoprotein (HEF) combines receptor binding, receptor hydrolysis and membrane fusion activities. Like other hemagglutinating glycoproteins of influenza viruses HEF is S‐acylated, but only with stearic acid at a single cysteine located at the cytosol‐facing end of the transmembrane region. Previous studies established the essential role of S‐acylation of hemagglutinin for replication of influenza A and B virus by affecting budding and/or membrane fusion, but the function of acylation of HEF was hitherto not investigated. Using reverse genetics we rescued a virus containing non‐stearoylated HEF, which was stable during serial passage and showed no competitive fitness defect, but the growth rate of the mutant virus was reduced by one log. Deacylation of HEF does neither affect the kinetics of its plasma membrane transport nor the protein composition of virus particles. Cryo‐electron microscopy showed that the shape of viral particles and the hexagonal array of spikes typical for influenza C virus were not influenced by this mutation indicating that virus budding was not disturbed. However, the extent and kinetics of haemolysis were reduced in mutant virus at 37°C, but not at 33°C, the optimal temperature for virus growth, suggesting that non‐acylated HEF has a defect in membrane fusion under suboptimal conditions.     

Influence of temperature, cholesterol, dipalmitoyllecithin and Ca2+ on the rate of muscle cell fusion

The rate of muscle cell fusion increases between 28 °C and 40 °C by a factor of 15 to 20. The formal activation energy of the fusion process changes abruptly at about 35 °C. This change is discussed in terms of a phase transition of the membrane lipids at 35 °C. In the presence of cholesterol or dipalmitoyllecithin the fusion rate decreases markedly. Increasing the temperature reverses the effects of cholesterol and dipalmitoyllecithin. These results are discussed in terms of interactions between membrane lipids.